A data cleaning method, device and equipment

Abstract

The application discloses a data cleaning method, device and equipment, and the method comprises the following steps: obtaining historical heat supply operation data of a heat exchange unit; if it is determined according to the historical heat supply operation data that target heat data in heat data under the same working condition is abnormal and other heat data is not all abnormal, then the abnormal target heat data is repaired by using the other heat data which is not abnormal under the same working condition; and if it is determined according to the historical heat supply operation data that the target heat data and the other heat data in the heat data under the same working condition are all abnormal, then the abnormal target heat data is repaired by using normal target heat data under other working conditions. According to the technical scheme disclosed by the application, various historical heat supply operation data are used to determine the abnormality of target heat data, and different repair methods are used for repair according to different abnormal conditions, so that the reliability and accuracy of the repair of abnormal target heat data are improved, and the accuracy of the load prediction of the heat exchange unit is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, more particularly, to a data cleaning method, device and equipment. BACKGROUND

[0002] The advent of the digital and intelligent era provides an opportunity for the development of heat supply technology. Among them, the application of load prediction technology brings great help to the fine function of heat supply enterprises. Real-time, accurate and stable measurement of heat data of heat exchange units is an important implementation basis for load prediction algorithms. However, due to monitoring equipment defects, human intervention and other reasons, the heat data of heat exchange units often contains a lot of missing values and abnormal values. The existence of these values greatly interferes with the analysis and calculation of the load prediction algorithm, resulting in relatively low accuracy of load prediction.

[0003] In summary, how to clean the data of heat exchange units in order to improve the accuracy of heat exchange unit load prediction is a technical problem that needs to be solved by the technical personnel in the field at present. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a data cleaning method, device and equipment for cleaning the data of heat exchange units in order to improve the accuracy of heat exchange unit load prediction.

[0005] In order to achieve the above purpose, the present application provides the following technical scheme:

[0006] A data cleaning method comprises:

[0007] Obtaining historical heat supply operation data of a heat exchange unit; the historical heat supply operation data comprises heat data and other heat supply operation data except the heat data;

[0008] If it is determined according to the historical heat supply operation data that the target heat data in the heat data under the same working condition is abnormal and other heat data is not all abnormal, then the abnormal target heat data is repaired using the other heat data under the same working condition which is not abnormal; the target heat data is heat data participating in load prediction;

[0009] If it is determined according to the historical heat supply operation data that the target heat data and other heat data in the heat data under the same working condition are all abnormal, then the abnormal target heat data is repaired using the normal target heat data under other working conditions.

[0010] Preferably, determining according to the historical heat supply operation data that the target heat data in the heat data under the same working condition is abnormal and other heat data is not all abnormal comprises:

[0011] determining a reasonable variation interval of the target heat data and a reasonable variation interval of other heat data by using a box plot or a 3σ principle;

[0012] if the target heat data under the same working condition is greater than or less than the reasonable variation interval of the target heat data, and other heat data under the same working condition is within the reasonable variation interval of the corresponding other heat data, it is determined that the target heat data among the heat data under the same working condition is abnormal, and the other heat data is not all abnormal;

[0013] repairing the abnormal target heat data by using the non-abnormal other heat data under the same working condition, comprising:

[0014] repairing the target heat data by using the other heat data with the best quality among the non-abnormal other heat data under the same working condition.

[0015] Preferably, according to the historical heat supply operation data, it is determined that the target heat data and the other heat data among the heat data under the same working condition are all abnormal, comprising:

[0016] when the heat exchange unit operates in a time-of-use adjustment mode and uses an electric valve adjustment to regulate heat, a reasonable variation interval of primary network electric valve opening degree feedback is determined by using a box plot or a 3σ principle, if the primary network electric valve opening degree feedback is not within the reasonable variation interval of the primary network electric valve opening degree feedback during the time-of-use adjustment, or the primary network electric valve opening degree feedback is within the reasonable variation interval of the primary network electric valve opening degree feedback during the time-of-use adjustment and is not within a reasonable variation sub-interval determined according to the reasonable variation interval of the primary network electric valve opening degree feedback, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0017] when the heat exchange unit operates in a time-of-use adjustment mode and uses a secondary network temperature adjustment to regulate heat, a reasonable variation interval of the secondary network temperature is determined by using a box plot or a 3σ principle, if at least one of the secondary network temperatures is not within the reasonable variation interval of the secondary network temperature during the time-of-use adjustment, or at least one of the secondary network temperatures is within the reasonable variation interval of the secondary network temperature during the time-of-use adjustment and is not within a reasonable variation sub-interval determined according to the reasonable variation interval of the secondary network temperature, it is determined that the corresponding target heat data and other heat data are all abnormal; wherein the secondary network temperature includes a secondary network water supply temperature setting, a secondary network return water temperature setting, and a secondary network supply and return water temperature setting;

[0018] repairing the abnormal target heat data by using normal target heat data under other working conditions, comprising:

[0019] The normal target heat data in the same adjustment period as the abnormal target heat data is processed by using a box plot or a 3σ principle to obtain a reasonable fluctuation range of the normal target heat data;

[0020] The abnormal target heat data is filled according to the reasonable fluctuation range of the normal target heat data.

[0021] Preferably, the target heat data and other heat data in the heat data under the same working condition are determined to be abnormal according to the historical heat supply operation data, and the method comprises the following steps:

[0022] The reasonable change intervals of the heat source instantaneous heat, the heat source instantaneous flow, the heat source water supply temperature, the primary network instantaneous heat and the primary network electric regulating valve opening degree feedback are determined by using a box plot or a 3σ principle; the heat transmission time length from the heat source to the analysis heat exchange unit is calculated by using the reasonable change intervals of the heat source instantaneous heat, the heat source instantaneous flow, the heat source water supply temperature and the primary network instantaneous heat; if the fluctuation of the heat source instantaneous heat exceeds the reasonable change subinterval determined according to the reasonable change interval of the heat source instantaneous heat, and the fluctuation of the heat source instantaneous heat reaches the analysis heat exchange unit after one heat transmission time length, and the primary network electric regulating valve opening degree feedback value does not exceed the reasonable change interval thereof, it is determined that the corresponding target heat data and other heat data are abnormal;

[0023] The reasonable change intervals of the primary network instantaneous heat, the primary network instantaneous flow, the primary network water supply temperature, the primary network return water temperature and the primary network electric regulating valve opening degree are determined by using a box plot or a 3σ principle; the heat transmission time length from the fluctuant heat exchange unit to the analysis heat exchange unit is calculated by using the reasonable change intervals of the primary network instantaneous heat, the primary network instantaneous flow, the primary network water supply temperature and the primary network return water temperature; if the fluctuation of the primary network instantaneous heat of the fluctuant heat exchange unit exceeds the reasonable change subinterval determined according to the reasonable change interval of the primary network instantaneous heat, and the fluctuation of the primary network instantaneous heat of the heat exchange unit reaches the analysis heat exchange unit after one heat transmission time length, and the primary network electric regulating valve opening degree feedback does not exceed the reasonable change interval thereof, it is determined that the corresponding target heat data and other heat data are abnormal;

[0024] The reasonable change intervals of the primary network electric regulating valve opening degree given, the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat are determined by using a box plot or a 3σ principle; the points, at which the primary network electric regulating valve opening degree given is not 0, the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat are all 0, are determined as critical points according to the reasonable change interval corresponding to the primary network electric regulating valve opening degree given, the reasonable change interval corresponding to the primary network electric regulating valve opening degree feedback and the reasonable change interval corresponding to the primary network instantaneous heat, and it is determined that the target heat data and other heat data below the critical points are abnormal.

[0025] analyzing the change of the instantaneous heat of the primary network in a certain time span when heat regulation is performed by adopting secondary network temperature regulation to obtain fluctuation change of the instantaneous heat of the primary network; when the peak amplitude and the trough amplitude in the fluctuation change of the instantaneous heat of the primary network are relatively consistent, the time interval from the peak to the trough is relatively consistent, and at least one of the peak amplitude or the trough amplitude is greater than a preset amplitude or the time interval from the peak to the trough is abnormal, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0026] If at least one of the following conditions occurs: the equipment data is abnormal, the frequency feedback value of the secondary network circulating water pump is 0, the feedback of the opening degree of the primary network electric regulating valve is normal, the secondary network water supply temperature deviates from its reasonable change interval, and the instantaneous heat of the primary network is 0, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0027] The reasonable change range of the feedback of the opening degree of the primary network electric regulating valve is determined by adopting a box plot or a 3σ principle, and when heat regulation is switched from being performed by adopting the feedback of the opening degree of the primary network electric regulating valve to being performed by adopting secondary network temperature regulation, if the feedback of the opening degree of the primary network electric regulating valve is not within the reasonable change range of the feedback of the opening degree of the primary network electric regulating valve, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0028] The abnormal target heat data is repaired by using normal target heat data under other working conditions, including:

[0029] If the number of abnormal target heat data is less than a preset value, the abnormal target heat data is filled by using the mean value of normal target heat data of adjacent points;

[0030] If the number of abnormal target heat data is not less than the preset value, the abnormal target heat data is filled by using normal target heat data within a preset time length before and a preset time length after.

[0031] Preferably, it is determined that the target heat data and other heat data in the heat data under the same working condition are all abnormal according to the historical heat supply operation data, including:

[0032] It is determined whether the target heat data and other heat data in the heat data under the same working condition are missing;

[0033] If the target heat data and other heat data under the same working condition are all missing, it is determined that the target heat data and other heat data in the heat data under the same working condition are all abnormal;

[0034] The abnormal target heat data is repaired by using normal target heat data under other working conditions, including:

[0035] When the heat exchange unit operates in the time-sharing adjustment mode, the normal target heat data in the same adjustment period as the abnormal target heat data is processed by using a box chart or a 3σ principle to obtain a reasonable fluctuation range of the normal target heat data; and the abnormal target heat data is filled according to the reasonable fluctuation range of the normal target heat data.

[0036] When the heat exchange unit operates in the time-sharing adjustment mode, if the number of abnormal target heat data is less than a preset value, the abnormal target heat data is filled by using the mean value of the normal target heat data of adjacent points; if the number of abnormal target heat data is not less than the preset value, the abnormal target heat data is filled by using the normal target heat data within a preset time length before and a preset time length after.

[0037] Preferably, according to the historical heat supply operation data, it is determined that the target heat data in the heat data under the same working condition is abnormal, and other heat data is not all abnormal, comprising:

[0038] It is determined whether the target heat data and other heat data in the heat data under the same working condition are missing.

[0039] If the target heat data under the same working condition is missing, and other heat data is not all missing, it is determined that the target heat data in the heat data under the same working condition is abnormal, and other heat data is not all abnormal.

[0040] The abnormal target heat data is repaired by using the non-abnormal other heat data under the same working condition, comprising:

[0041] The target heat data is repaired by using the other heat data with the best quality in the non-abnormal other heat data under the same working condition.

[0042] Preferably, the method further comprises:

[0043] The time parameters corresponding to the historical heat supply operation data and the building properties corresponding to the heat exchange unit are obtained; the time parameters are divided into weekdays and weekends.

[0044] According to the target heat data of the heat exchange unit corresponding to different building properties at different time parameters, the change rule of the target heat data of the corresponding heat exchange unit at different time parameters is determined.

[0045] If the target heat data of the heat exchange unit corresponding to the building property exceeds the target heat data variation rule of the heat exchange unit corresponding to the building property at the corresponding time parameter in the analysis period, the normal target heat data of the heat exchange unit corresponding to the building property at the corresponding time parameter and other working conditions is used to repair the abnormal target heat data.

[0046] Preferably, further comprising:

[0047] Obtaining the outdoor temperature corresponding to the historical heating operation data;

[0048] According to the target heat data corresponding to the difference of the outdoor temperature less than the preset temperature difference value, the corresponding target heat data variation interval is determined.

[0049] If the target heat data corresponding to the outdoor temperature exceeds the corresponding target heat data variation interval, the normal target heat data under other working conditions is used to repair the abnormal target heat data.

[0050] A data cleaning device, comprising:

[0051] A first obtaining module is configured to obtain historical heating operation data of a heat exchange unit; the historical heating operation data comprises heat data and other heating operation data except the heat data;

[0052] A first repairing module is configured to, if it is determined according to the historical heating operation data that the target heat data in the heat data under the same working condition is abnormal and other heat data is not all abnormal, use the normal heat data under the same working condition to repair the abnormal target heat data; the target heat data is the heat data participating in load prediction.

[0053] A second repairing module is configured to, if it is determined according to the historical heating operation data that the target heat data and other heat data in the heat data under the same working condition are all abnormal, use the normal target heat data under other working conditions to repair the abnormal target heat data.

[0054] A data cleaning device, comprising:

[0055] A memory is configured to store a computer program;

[0056] A processor is configured to execute the computer program to realize the steps of the data cleaning method according to any one of the above.

[0057] The application provides a data cleaning method, device and equipment, wherein the method comprises the following steps: obtaining historical heat supply operation data of a heat exchange unit; the historical heat supply operation data comprises heat data and other heat supply operation data except the heat data; if it is determined according to the historical heat supply operation data that target heat data in the heat data under the same working condition is abnormal and other heat data under the same working condition is not all abnormal, the abnormal target heat data is repaired by using the normal other heat data under the same working condition; the target heat data is heat data participating in load prediction; if it is determined according to the historical heat supply operation data that the target heat data and other heat data in the heat data under the same working condition are all abnormal, the abnormal target heat data is repaired by using normal target heat data under other working conditions.

[0058] The above technical solution disclosed by the application obtains the historical heat supply operation data of the heat exchange unit, and the historical heat supply operation data comprises heat data and other heat supply operation data; when it is determined according to the obtained historical heat supply operation data that the target heat data in the heat data under the same working condition is abnormal and other heat data under the same working condition is not all abnormal, the abnormal target heat data is repaired by using the normal other heat data under the same working condition; when it is determined according to the obtained historical heat supply operation data that the target heat data and other heat data in the heat data under the same working condition are all abnormal, the abnormal target heat data is repaired by using normal target heat data under other working conditions. Through the above process, the abnormality of the target heat data can be determined by using various historical heat supply operation data of the heat exchange unit, so that the target heat data can be comprehensively analyzed, the abnormal target heat data is avoided to be missed, and the accuracy of the abnormality determination of the target heat data is improved. Moreover, through the above process, different repair methods can be used to repair the abnormal target heat data according to different abnormal conditions, so that the reliability and accuracy of the repair of the abnormal target heat data are improved, and the repaired target heat data can better participate in the load prediction algorithm, thereby improving the accuracy of the load prediction of the heat exchange unit. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of the provided drawings.

[0060] Figure 1 A flowchart of a data cleaning method provided by the embodiment of the present application;

[0061] Figure 2 A flowchart of another data cleaning method provided by the embodiment of the present application;

[0062] Figure 3 A structural schematic diagram of a data cleaning device provided by an embodiment of the present application is provided.

[0063] Figure 4 A structural schematic diagram of a data cleaning device provided by an embodiment of the present application is provided. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0065] Referring to Figure 1 which shows a flowchart of a data cleaning method provided by an embodiment of the present application. The data cleaning method provided by an embodiment of the present application can include:

[0066] S11: obtaining historical heat supply operation data of a heat exchange unit; the historical heat supply operation data includes heat data and other heat supply operation data except heat data.

[0067] In the present application, first, historical heat supply operation data of a heat exchange unit can be obtained, specifically, heat supply operation data of the heat exchange unit in a preset historical period (which can be day, week, ten days, month, season, heating season, etc.). The historical heat supply operation data obtained by the present application contains other heat supply operation data except heat data and heat data, and the heat data includes primary network heat data and secondary network heat data.

[0068] Specifically, the heat data in the heat data provided by the present application can specifically include primary network instantaneous heat, primary network calculated heat, primary network cumulative heat difference, secondary network instantaneous heat, secondary network calculated heat and secondary network cumulative heat difference. The above six parameters have the following calculation relationship:

[0069]

[0070]

[0071] Qljc=Qdlj-Qslj——Formula 3

[0072] Qss1 is the instantaneous heat of the primary network, Qjs1 is the calculated heat of the primary network, Qljc1 is the cumulative heat difference of the primary network, Qss2 is the instantaneous heat of the secondary network, Qjs2 is the calculated heat of the secondary network, Qljc2 is the cumulative heat difference of the secondary network, η is the heat exchange efficiency of the heat exchanger, Qjs is the calculated heat of the primary network or the secondary network, G is the instantaneous flow of the primary network or the secondary network, c is the specific heat capacity of water, tg is the supply water temperature of the primary network or the secondary network, th is the return water temperature of the primary network or the secondary network, Qljc is the cumulative heat difference of the primary network or the secondary network, Qdlj is the cumulative heat at the current integral point of the primary network or the secondary network, Qslj is the cumulative heat at the last integral point of the primary network or the secondary network.

[0073] The other heating operation data can specifically include heat source instantaneous heat, heat source instantaneous flow, heat source supply water temperature, primary network cumulative heat, primary network supply water temperature, primary network return water temperature, primary network instantaneous flow, primary network electric regulating valve opening degree given, primary network electric regulating valve opening degree feedback, primary network electric regulating valve hand-automatic switching, secondary network cumulative heat, secondary network supply water temperature, secondary network return water temperature, secondary network supply water temperature setting, secondary network return water temperature setting, secondary network supply-return water temperature setting, secondary network instantaneous flow, secondary network circulating water pump frequency given, secondary network circulating water pump frequency feedback, communication state, cumulative electricity consumption.

[0074] S12: If it is determined according to the historical heating operation data that the target heat data in the heat data under the same working condition is abnormal and other heat data is not all abnormal, the abnormal target heat data is repaired by using the other heat data under the same working condition which is not abnormal; the target heat data is the heat data participating in load prediction.

[0075] On the basis of step S11, whether the target heat data and other quantity data (i.e. heat data other than the target heat data) in the heat data under the same condition are all abnormal can be analyzed and judged according to the obtained historical heating operation data. The same working condition mentioned here is specifically the same time point, the target heat data is the heat data participating in load prediction, for example, can be the primary network instantaneous heat or the secondary network instantaneous heat, and the abnormal condition can be data missing or data not missing but incorrect. In this way, it is realized that whether the target heat data is abnormal is determined by using the historical heating operation data, that is, not only the heat data in the historical heating operation data is used to determine whether the target heat data is abnormal, but also other heating operation data is used to determine whether the target heat data is abnormal, so as to improve the accuracy of the determination of the abnormal target heat data, avoid missing the abnormal target heat data, thereby facilitating accurate cleaning of the target heat data, and further facilitating improvement of the accuracy of load prediction.

[0076] When the target heat data in the heat data under the same working condition is determined to be abnormal according to the historical heat supply operation data, and other heat data is not all abnormal, the abnormal target heat data under the same working condition can be repaired by using the non-abnormal other heat data belonging to the same working condition, that is, when other heat data under the same working condition is not all abnormal, the target heat data is repaired by using other heat data under the same working condition, so as to improve the reliability and accuracy of the target heat data cleaning, and then the accuracy of the load prediction can be improved when the target heat data is used in the load prediction.

[0077] S13: If the target heat data and other heat data in the heat data under the same working condition are determined to be abnormal according to the historical heat supply operation data, the abnormal target heat data is repaired by using normal target heat data under other working conditions.

[0078] When the target heat data and other heat data in the heat data under the same working condition are determined to be abnormal according to the historical heat supply operation data, at this time, the abnormal target heat data can be repaired by using normal target heat data under other working conditions, which can be specifically analyzed, and the corresponding and appropriate data corresponding to the abnormal target heat data under the above working condition is obtained, and then the above working condition is filled, so as to improve the reliability and accuracy of the abnormal target heat data repair, and then the accuracy of the load prediction can be improved when the target heat data is used in the load prediction.

[0079] The above technical solution disclosed in the application obtains the historical heat supply operation data of the heat exchange unit containing heat data and other heat supply operation data, when the target heat data in the heat data under the same working condition is determined to be abnormal according to the obtained historical heat supply operation data, and other heat data is not all abnormal, the abnormal target heat data is repaired by using non-abnormal other heat data under the same working condition, when the target heat data and other heat data in the heat data under the same working condition are determined to be abnormal according to the obtained historical heat supply operation data, the abnormal target heat data is repaired by using normal target heat data under other working conditions. Through the above process, the abnormality of the target heat data can be determined by using various historical heat supply operation data of the heat exchange unit, so as to realize comprehensive analysis of the target heat data, avoid missing abnormal target heat data, improve the accuracy of the target heat data abnormality determination. And through the above process, different repair methods can be used to repair the abnormal target heat data according to different abnormal conditions, so as to improve the reliability and accuracy of the abnormal target heat data repair, so that the repaired target heat data can better participate in the load prediction algorithm, and then the accuracy of the load prediction of the heat exchange unit is improved.

[0080] See Figure 2 The diagram illustrates a flowchart of another data cleaning method provided in an embodiment of this application. This data cleaning method, based on historical heating operation data, determines whether target heat data is abnormal and other heat data are not all abnormal within the same operating condition. It may include:

[0081] Use box plots or the 3σ principle to determine the reasonable variation range of the target heat data and the reasonable variation range of other heat data;

[0082] If the target heat data under the same working condition is greater than or less than the reasonable variation range of the target heat data, and there are other heat data under the same working condition that are within the reasonable variation range of the corresponding other heat data, then it is determined that the target heat data under the same working condition is abnormal, and the other heat data are not all abnormal.

[0083] Repairing abnormal target heat data using other heat data that are not abnormal under the same operating conditions can include:

[0084] The target heat data is repaired by using the best quality heat data from other heat data that are not abnormal under the same operating conditions.

[0085] In this application, box plots or the 3σ principle can be used to determine the reasonable variation range of the target heat data and the reasonable variation range of other heat data. If the target heat data under the same operating condition is greater than or less than the reasonable variation range of the target heat data (i.e., the target heat data is not within its reasonable variation range), while other heat data under the same operating condition are within their corresponding reasonable variation ranges, then it can be determined that the target heat data under the same operating condition is abnormal, and not all other heat data are abnormal. Then, the above three formulas can be used to cross-verify the groups of other heat data that are not abnormal (the groups mentioned here are data groups composed of the same type of data under different operating conditions) using historical heating operation data within a certain time span (i.e., the preset historical period mentioned above). From these groups, a group of other heat data with high stability and accuracy can be selected (for example, the group can be the one with the highest percentage of normal other heat data). The abnormal target heat data can then be replaced and filled using other heat data in the same operating condition as the abnormal target heat data (i.e., the other heat data with the best quality). Alternatively, several other sets of heat data can be selected, and the average of the other heat data belonging to the same working condition as the abnormal target heat data in these sets of heat data can be calculated to obtain the other heat data with the best quality. Then, the abnormal target heat data under the same working condition can be replaced and filled.

[0086] Through the above manner, the reliability and accuracy of the abnormal target heat data processing can be improved, thereby improving the quality of the target heat data cleaning, and further facilitating the accuracy of the load prediction.

[0087] In addition, in the present application, the target heat data can also be determined according to the communication state of the historical heat supply operation data. Specifically, if the communication state is interrupted and the target heat data of a certain period is not recorded, it can be determined that the target heat data is abnormal. At this time, the target heat data of the period can be imported from the local platform system to fill in and repair the unrecorded target heat data by using the imported target heat data.

[0088] The data cleaning method provided by the embodiment of the present application can include the following steps:

[0089] When the heat exchange unit operates in a time period adjustment mode and adopts an electric valve adjustment to regulate heat, a box chart or a 3σ principle is used to determine a reasonable change interval of the primary network electric valve opening degree feedback. If the primary network electric valve opening degree feedback is not in the reasonable change interval of the primary network electric valve opening degree feedback during the time period adjustment, or the primary network electric valve opening degree feedback is in the reasonable change interval of the primary network electric valve opening degree feedback during the time period adjustment and is not in a reasonable change sub-interval determined according to the reasonable change interval of the primary network electric valve opening degree feedback, it is determined that the corresponding target heat data and other heat data are abnormal.

[0090] When the heat exchange unit operates in a time period adjustment mode and adopts a secondary network temperature adjustment to regulate heat, a box chart or a 3σ principle is used to determine a reasonable change interval of the secondary network temperature. If at least one of the secondary network temperatures is not in the reasonable change interval of the secondary network temperature during the time period adjustment, or at least one of the secondary network temperatures is in the reasonable change interval of the secondary network temperature during the time period adjustment and is not in a reasonable change sub-interval determined according to the reasonable change interval of the secondary network temperature, it is determined that the corresponding target heat data and other heat data are abnormal. The secondary network temperature can include a secondary network water supply temperature setting, a secondary network return water temperature setting, and a secondary network water supply and return temperature setting.

[0091] The normal target heat data in the same adjustment period as the abnormal target heat data is processed by using a box chart or a 3σ principle to obtain a reasonable fluctuation range of the normal target heat data.

[0092] The normal target heat data in the same adjustment period as the abnormal target heat data is processed by using a box chart or a 3σ principle to obtain a reasonable fluctuation range of the normal target heat data.

[0093] The abnormal target heat data is filled according to a reasonable fluctuation range of normal target heat data.

[0094] In the present application, when the heat exchange unit is running in the regulation period of the time-sharing regulation mode (in the time-sharing regulation mode, a certain operation mode can last for a time period, that is, the operation mode at each time point is the same within a time period), for the regulation mode of heat regulation by electric valve regulation, the once network electric regulating valve opening degree feedback in the obtained historical heating operation data can be analyzed, the box chart or 3σ principle is used to determine the reasonable change interval of the once network electric regulating valve opening degree feedback, and the reasonable change sub-interval of the once network electric regulating valve opening degree feedback can be determined according to the reasonable change interval of the once network electric regulating valve opening degree feedback (that is, a reasonable change sub-interval is determined according to the fluctuation of the once network electric regulating valve opening degree feedback in the reasonable change interval). If the once network electric regulating valve opening degree feedback is not in the reasonable change interval of the once network electric regulating valve opening degree feedback during the time-sharing regulation period, or the once network electric regulating valve opening degree feedback is in its reasonable change interval but not in its reasonable change sub-interval during the time-sharing regulation period, it is determined that the target heat data and other heat data corresponding to the above-mentioned situation are abnormal. Of course, for the above-mentioned regulation mode, the data difference of the two parameters of the obtained historical heating operation data, that is, the once network electric regulating valve opening degree given (generally a fixed value) and the once network electric regulating valve opening degree feedback can be analyzed, the box chart or 3σ principle is used to determine the reasonable change interval of the data difference, and the reasonable change sub-interval is obtained according to the reasonable change interval of the data difference, if the data difference of the two is not in the reasonable change interval of the once network electric regulating valve opening degree feedback during the time-sharing regulation period, or the data difference of the two is in its reasonable change interval but not in its reasonable change sub-interval during the time-sharing regulation period, it is determined that the corresponding target heat data and other heat data are abnormal. For the above-mentioned situation, the following is taken as an example: when the time-sharing regulation mode is used for regulation, due to the self-reasons of the once network side electric regulating valve actuator, in the process of executing opening degree regulation, it cannot directly reach the given value, but will first open to a larger or smaller value along the trend of the execution value, and then return to the given value, resulting in abnormal target quantity parameters during the regulation period.

[0095] When the heat exchange unit operates in the regulation period of the split-time regulation mode, for the heat regulation mode of regulating heat by using the secondary network temperature, the reasonable variation interval of the secondary network temperature can be determined by using the box plot or the 3σ principle, and the reasonable variation sub-interval can be determined according to the reasonable variation interval. The secondary network temperature mentioned herein specifically includes the secondary network supply water temperature setting, the secondary network return water temperature setting, and the secondary network supply and return water temperature setting. After the reasonable variation interval of each secondary network temperature is determined, if at least one of the secondary network temperatures is not in the reasonable variation interval of the secondary network temperature during the split-time regulation period, or at least one of the secondary network temperatures is in the reasonable variation interval of the secondary network temperature but not in the corresponding reasonable variation sub-interval during the split-time regulation period, it is determined that the corresponding target heat data and other heat data are abnormal.

[0096] For the above two cases, the target heat data can be repaired in the following manner: first, the normal target heat data in the same regulation period as the abnormal target heat data is processed by using the box plot or the 3σ principle to obtain the reasonable fluctuation range of the normal target heat data; then, the abnormal target heat data is filled (i.e., replaced) according to the mean, median or mode of the reasonable fluctuation range of the normal target heat data.

[0097] By the above-mentioned manner, the target heat data is determined to be abnormal according to different regulation modes by using different other heat data, so as to improve the reliability and accuracy of the target heat data abnormality determination, and the abnormal target heat data is repaired according to the corresponding repair method, so as to improve the reliability of the target heat data repair, and further facilitate to improve the accuracy of the heat exchange unit load prediction.

[0098] The data cleaning method provided by the embodiment of the present application can determine that the target heat data and other heat data in the heat data under the same working condition are all abnormal according to the historical heat supply operation data, which can include:

[0099] The reasonable variation intervals of the heat source instantaneous heat, the heat source instantaneous flow, the heat source supply water temperature, the primary network instantaneous heat, and the primary network electric regulating valve opening degree feedback are determined by using the box plot or the 3σ principle; the heat transfer time from the heat source to the analysis heat exchange unit is calculated by using the reasonable variation intervals of the heat source instantaneous heat, the heat source instantaneous flow, the heat source supply water temperature, and the primary network instantaneous heat; if the fluctuation of the heat source instantaneous heat exceeds the reasonable variation sub-interval determined according to the reasonable variation interval of the heat source instantaneous heat, and the fluctuation of the heat source instantaneous heat reaches the analysis heat exchange unit after one heat transfer time, and the primary network electric regulating valve opening degree feedback value does not exceed the reasonable variation interval, it is determined that the corresponding target heat data and other heat data are all abnormal.

[0100] Determine the reasonable variation interval of the primary network instantaneous heat, the primary network instantaneous flow, the primary network water supply temperature, the primary network return water temperature and the primary network electric regulating valve opening degree respectively by using the box plot or the 3σ principle; calculate the heat transfer duration from the fluctuation heat exchange unit to the analysis heat exchange unit by using the reasonable variation interval of the primary network instantaneous heat, the reasonable variation interval of the primary network instantaneous flow, the reasonable variation interval of the primary network water supply temperature and the reasonable variation interval of the primary network return water temperature; if the fluctuation of the primary network instantaneous heat of the fluctuation heat exchange unit exceeds the reasonable variation subinterval determined according to the reasonable variation interval of the primary network instantaneous heat, and the fluctuation of the primary network instantaneous heat of the heat exchange unit reaches the analysis heat exchange unit after one heat transfer duration, and the feedback of the primary network electric regulating valve opening degree does not exceed the reasonable variation interval, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0101] Determine the reasonable variation interval of the primary network electric regulating valve opening degree given, the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat respectively by using the box plot or the 3σ principle; take the point that the primary network electric regulating valve opening degree given is not 0, the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat are all 0 as the critical point according to the reasonable variation interval of the primary network electric regulating valve opening degree given, the reasonable variation interval of the primary network electric regulating valve opening degree feedback and the reasonable variation interval of the primary network instantaneous heat, and determine that the target heat data and other heat data below the critical point are all abnormal;

[0102] When the heat is regulated by using the secondary network temperature regulation, analyze the variation of the primary network instantaneous heat within a certain time span to obtain the fluctuation variation of the primary network instantaneous heat; when the peak amplitude and the valley amplitude in the fluctuation variation of the primary network instantaneous heat are relatively consistent, the time interval from the peak to the valley is relatively consistent, and there is at least one of the peak amplitude or the valley amplitude greater than the preset amplitude or the time interval from the peak to the valley abnormal, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0103] If at least one of the following conditions occurs: the equipment data is abnormal, the secondary network circulating water pump frequency feedback value is 0, the primary network electric regulating valve opening degree feedback is normal, the secondary network water supply temperature deviates from its reasonable variation interval, and the primary network instantaneous heat is 0, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0104] Determine the reasonable variation range of the primary network electric regulating valve opening degree feedback by using the box plot or the 3σ principle; when switching from regulating the heat by using the primary network electric regulating valve opening degree feedback to regulating the heat by using the secondary network temperature regulation, if the primary network electric regulating valve opening degree feedback is not within the reasonable variation range of the primary network electric regulating valve opening degree feedback, it is determined that the corresponding target heat data and other heat data are all abnormal;

[0105] Repairing the abnormal target heat data with normal target heat data under other working conditions can include:

[0106] If the number of abnormal target heat data is less than the preset value, the abnormal target heat data is filled with the mean value of normal target heat data of adjacent points.

[0107] If the number of abnormal target heat data is not less than the preset value, the abnormal target heat data is filled with normal target heat data within the preset time length before and after.

[0108] The change of heat source heating parameters will also cause abnormal target heat parameters and other heat parameters. Therefore, the application first determines the reasonable change interval of the heat source instantaneous heat, the heat source instantaneous flow, the heat source water temperature, the primary network instantaneous heat and the primary network electric valve opening degree feedback in the historical heating operation data obtained by using the box chart or the 3σ principle. Then, the heat transfer time length of the heat source to the analysis heat exchange unit can be calculated by using the reasonable change interval of the heat source instantaneous heat, the reasonable change interval of the heat source instantaneous flow, the reasonable change interval of the heat source water temperature and the reasonable change interval of the primary network instantaneous heat. After that, if the fluctuation of the heat source instantaneous heat exceeds the reasonable change subinterval determined according to the reasonable change interval of the heat source instantaneous heat, the fluctuation reaches the analysis heat exchange unit after one heat transfer time length, and the primary network electric valve opening degree feedback does not exceed its reasonable change interval (i.e. excluding the influence of the primary network electric valve opening degree feedback on the target heat data), it can be determined that the target heat data and other heat data are abnormal, and the abnormality is caused by the change of the heat source heating parameters. For the above situation, for example: the number of boilers on the heat source side is increased or decreased, and when the heat fluctuation reaches the heat exchange unit, the primary network electric valve does not timely track the change, resulting in abnormal heat data in local period.

[0109] In addition, the change of the heating parameter of the adjacent heat exchange unit can also cause the target heat parameter and other heat parameters to be abnormal. Therefore, the application first determines the reasonable change interval of the primary network instantaneous heat, the primary network instantaneous flow, the primary network water supply temperature, the primary network return water temperature, and the primary network electric regulating valve opening degree corresponding to the historical heating operation data obtained by using a box plot or a 3σ principle. Then, the heat transfer time length from the fluctuating heat exchange unit to the analysis heat exchange unit can be calculated using the reasonable change interval of the primary network instantaneous heat, the reasonable change interval of the primary network instantaneous flow, the reasonable change interval of the primary network water supply temperature, and the reasonable change interval of the primary network return water temperature. After that, if the fluctuation of the primary network instantaneous heat of the fluctuating heat exchange unit exceeds the reasonable change subinterval determined according to the reasonable change interval of the primary network instantaneous heat, and the fluctuation of the primary network instantaneous heat of the heat exchange unit reaches the analysis heat exchange unit after one heat transfer time length, and the feedback of the primary network electric regulating valve opening degree does not exceed its reasonable change interval (i.e. excluding the influence of the feedback of the primary network electric regulating valve opening degree on the target heat data), it is determined that the corresponding target heat data and other heat data are abnormal, and it can be determined that the abnormality is caused by the change of the heating parameter of the adjacent heat exchange unit. For the above-mentioned situation, for example, the maintenance and power failure of the adjacent heat exchange unit cause the water force fluctuation of the heating pipe network system.

[0110] In addition, the characteristics of the equipment itself and the switching of the control mode can also cause the target heat parameter and other heat parameters to be abnormal. Therefore, the application processes the following four cases:

[0111] Case 1: The reasonable change interval of the primary network electric regulating valve opening degree given, the primary network electric regulating valve opening degree feedback, and the primary network instantaneous heat corresponding to the historical heating operation data obtained is determined by using a box plot or a 3σ principle. Then, the data is extracted from the reasonable change interval of the above-mentioned three parameters, and it is analyzed that the primary network electric regulating valve opening degree is not 0, but the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat are both 0, and the point is taken as a critical point, that is, the point where the primary network electric regulating valve opening degree given is not 0, the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat are both 0 is taken as a critical point. After that, the target heat data and other heat data below the critical point are all determined as abnormal heat data, that is, it is determined that the target heat data and other heat data below the critical point are abnormal. For the foregoing case, for example, due to the high outdoor temperature, the demand heat of the heat user is very small, at this time, the primary network electric regulating valve opening degree given value is a very small value, due to the characteristics of the primary network electric regulating valve, the difference between the primary network electric regulating valve opening degree given value and the feedback cannot maintain the original reasonable change interval, the difference becomes 0, causing the target heat data and other heat data to be 0.

[0112] Case 2: When heat regulation is performed by using secondary network temperature (which can include secondary network water supply temperature setting, secondary network return water temperature setting, secondary network supply and return water temperature setting), the change of primary network instantaneous heat in a certain time span in the obtained historical heating operation data is analyzed to obtain the fluctuation change of the primary network instantaneous heat. Then, the fluctuation change of the primary network instantaneous heat is counted. If the peak amplitude and the valley amplitude in the fluctuation change of the primary network instantaneous heat are relatively consistent (i.e., the peak amplitude and the valley amplitude differ by less than a first threshold value), the time interval from the peak to the valley is relatively consistent (i.e., the time interval from the adjacent peak to the valley and the time interval from another adjacent peak to the valley differ by less than a second threshold value), but in the fluctuation change of the primary network instantaneous heat, there is a peak amplitude or a valley amplitude greater than a preset amplitude, or the time interval from the peak to the valley is abnormal, that is, in the fluctuation change of the primary network instantaneous heat, most (the proportion of the peak amplitude and the valley amplitude being relatively consistent exceeds a first set proportion) of the peak amplitude and the valley amplitude are relatively consistent, most of the time interval from the peak to the valley is relatively consistent (the proportion of the time interval from the peak to the valley being relatively consistent exceeds a second set proportion), but in the fluctuation change, there is a peak amplitude or a valley amplitude not within the above-mentioned corresponding relatively consistent range, or the time interval from the peak to the valley is not within the above-mentioned corresponding relatively consistent range, then it can be determined that the corresponding target heat data and other heat data are all abnormal. For the above case, an example is as follows: due to unreasonable PID parameter setting of the primary network electric regulating valve, the opening degree thereof jumps up and down around the instantaneous heat value corresponding to a certain parameter setting value with equal amplitude.

[0113] Case 3: If at least one of the following conditions is met according to the obtained historical heating operation data: device data anomaly occurs (specifically, device data is abnormal without reason, and these data can be specifically data extracted from a heat exchange station), the secondary network circulating water pump frequency feedback value is 0, the primary network electric regulating valve opening degree feedback is normal, the secondary network water supply temperature deviates from its reasonable change interval, and the primary network instantaneous heat is 0, then it is determined that the corresponding target heat data and other heat data are all abnormal. For the above case, examples are as follows: ① device data upload anomaly; ② heat exchange unit secondary network pump stop, resulting in rapid increase of secondary network water supply temperature and decrease of instantaneous heat value; ③ heat exchange unit shutdown for maintenance, primary network instantaneous flow is 0, and instantaneous heat value is 0.

[0114] Case 4: The reasonable change range of the primary net electric regulating valve opening degree feedback is determined by using a box chart or a 3σ principle, when switching from the heat regulation by using the primary net electric regulating valve opening degree feedback to the heat regulation by using the secondary net temperature regulation (i.e. when executing the primary net electric regulating valve hand automatic switching command), if the primary net electric regulating valve opening degree feedback is not within the reasonable change range of the primary net electric regulating valve opening degree feedback, it is determined that the corresponding target heat data and other heat data are abnormal. For the foregoing case, the following is taken as an example: when the primary net electric regulating valve hand automatic switching, due to the change of the target value tracked and the influence of the electric regulating valve actuator itself characteristics, in the process of executing the opening degree regulation, it cannot directly reach the given value, but will first open to a larger or smaller value along the trend of the execution value, and then return to the given value, resulting in abnormal instantaneous heat value during the regulation.

[0115] For each of the above-mentioned heat source heating parameter changes, the adjacent heat exchange unit heating parameter changes, the equipment itself characteristics and the control mode switching, after determining that the corresponding target heat data and other heat data are all abnormal, when filling in the abnormal target heat data, the number of abnormal target heat data can be determined and compared with a preset value (previously set according to actual needs). If the number of abnormal target heat data is less than the preset value, the mean value of the normal target heat data of the adjacent point can be used to fill in the abnormal target heat data; if the number of abnormal target heat data is not less than the preset value, the normal target heat data within the preset time length (for example, the previous 7 days, the previous 10 days, the previous 1 month) and the normal target heat data within the preset time length (correspondingly, the next 7 days, the next 10 days, the next 1 month) are used to fill in the abnormal target heat data, that is, the normal target heat data within the double preset time length (the preset time length before the occurrence of the abnormal target heat data and the preset time length after the occurrence of the abnormal target heat data) is used to fill in the abnormal target heat data. Specifically, the normal target heat data within the preset time length and the normal target heat data within the preset time length can be used to build an algorithm model (such as an SVR model), and the algorithm model is used to fill in the abnormal target heat data, or the control period with the highest frequency of target heat data change law can be obtained from the preset time length and the preset time length, and the target heat data of the control period is used to fill in the abnormal target heat data. For case 1 in the equipment itself characteristics and the control mode switching, when the number of abnormal target heat data is too large, the critical target heat data when the instantaneous heat changes to 0 value in the adjacent time interval (7 days, 10 days, 1 month, etc.) of the abnormal period can be used for filling and repairing, and the specific filling and repairing method is the same as the above-mentioned specific method for filling and repairing when the number of abnormal target heat data is greater than the preset value, except that the critical target heat data when the instantaneous heat changes to 0 value is used for processing.

[0116] By the above-mentioned method, different other heating operation data can be used to determine the heat data abnormality for different cases, so as to improve the accuracy of heat data abnormality determination, and the target heat data can be repaired and filled in a corresponding way according to the number of abnormal target heat data, so as to improve the accuracy of target heat data repair and filling.

[0117] The data cleaning method provided by the embodiment of the application can determine that the target heat data and other heat data in the heat data under the same working condition are all abnormal according to historical heating operation data, which can include:

[0118] determining whether the target heat data and the other heat data in the heat data under the same working condition are missing;

[0119] if the target heat data and the other heat data under the same working condition are all missing, determining that the target heat data and the other heat data in the heat data under the same working condition are abnormal;

[0120] repairing the abnormal target heat data by using the normal target heat data under other working conditions, which can include:

[0121] when the heat exchange unit is operated in a split-time period adjustment mode, the normal target heat data in the same adjustment time period as the abnormal target heat data is processed by using a box chart or a 3σ principle to obtain a reasonable fluctuation range of the normal target heat data, and the abnormal target heat data is filled according to the reasonable fluctuation range of the normal target heat data;

[0122] when the heat exchange unit is operated in a time-by-time adjustment mode, if the number of the abnormal target heat data is less than a preset value, the abnormal target heat data is filled by using the mean value of the normal target heat data of adjacent points; if the number of the abnormal target heat data is not less than the preset value, the abnormal target heat data is filled by using the normal target heat data within a preset time length before and a preset time length after.

[0123] In the present application, the target heat data and other heat data in the heat data under the same working condition can also be determined directly from the target heat data and other heat data contained in the acquired historical heat data. If the target heat data and other heat data under the same working condition are all missing, it is determined that the target heat data and other heat data in the heat data under the same working condition are abnormal. For this case, when repairing the target heat data, if the heat exchange unit is running in a time-of-day adjustment mode, the normal target heat data in the same adjustment period as the abnormal target heat data can be processed by using a box plot or a 3σ principle to obtain a reasonable fluctuation range of the normal target heat data, and then the abnormal target heat data is filled according to the reasonable fluctuation range of the normal target heat data. The specific processing process is similar to the above-mentioned corresponding processing process, and will not be repeated here. If the heat exchange unit is running in a time-by-time adjustment mode, if the number of abnormal target heat data is less than a preset value, the mean value of the normal target heat data of adjacent points is used to fill the abnormal target heat data; if the number of abnormal target heat data is not less than the preset value, the normal target heat data within a preset time length before and the normal target heat data within a preset time length after are used to fill the abnormal target heat data. The specific processing process is similar to the above-mentioned corresponding processing process, and will not be repeated here.

[0124] By the above-mentioned manner, the determination of the missing target heat data and other heat data can be realized, and different ways are used to repair and fill the missing target heat data for different adjustment modes, so as to improve the accuracy of target heat data cleaning.

[0125] The data cleaning method provided by the embodiment of the present application can determine that the target heat data is abnormal and the other heat data is not all abnormal in the heat data under the same working condition according to the historical heat supply operation data, which can include:

[0126] It is determined whether the target heat data and other heat data in the heat data under the same working condition are missing;

[0127] If the target heat data under the same working condition is missing and the other heat data is not all missing, it is determined that the target heat data in the heat data under the same working condition is abnormal and the other heat data is not all abnormal;

[0128] The abnormal target heat data is repaired by using the non-abnormal other heat data under the same working condition, which can include:

[0129] The target heat data is repaired by using the other heat data with the best quality in the non-abnormal other heat data under the same working condition.

[0130] In the present application, whether the target heat data and the other heat data in the heat data under the same working condition are missing is determined from the target heat data and the other heat data contained in the heat data obtained from the historical heat data. If the target heat data under the same working condition is missing and the other heat data is not all missing, it is determined that the target heat data in the heat data under the same working condition is abnormal and the other heat data is not all abnormal. For this case, when repairing the target heat data, the target heat data can be repaired by using the other heat data of the best quality in the other heat data under the same working condition which is not abnormal, which is similar to the above-mentioned corresponding processing process, and will not be described here.

[0131] By the above-mentioned manner, the determination of the missing of the target heat data and the other heat data can be realized, and the target heat data is repaired by using the other heat data of the best quality under the same working condition which is not abnormal, so as to improve the reliability and accuracy of the target heat data repair.

[0132] The data cleaning method provided by the embodiment of the present application can further include:

[0133] The time parameters corresponding to the historical heat supply operation data and the building properties corresponding to the heat exchange units are obtained; the time parameters are divided into weekdays and weekends;

[0134] The target heat data variation law of the heat exchange units corresponding to different building properties under different time parameters is determined according to the target heat data of the heat exchange units corresponding to different building properties under different time parameters;

[0135] If the target heat data of the heat exchange units corresponding to the building properties in the analysis period exceeds the target heat data variation law of the heat exchange units corresponding to the building properties under the corresponding time parameters, the abnormal target heat data is repaired by using the normal target heat data of the heat exchange units corresponding to the building properties under the corresponding time parameters and other working conditions.

[0136] In the present application, while obtaining the historical heat supply operation data of the heat exchange units, the time parameters corresponding to the historical heat supply operation data and the building properties corresponding to the heat exchange units are also obtained, wherein the time parameters are divided into weekdays and weekends, and the building properties can specifically refer to public buildings, residential buildings, or more subdivided categories. In addition, the user heat characteristics, i.e. the different heat habits of users, can also be obtained, for example: office buildings: the daytime heat consumption is large on weekdays, the nighttime heat consumption is small, and the diurnal heat consumption is small on weekends; residential buildings: the daytime heat consumption is small on weekdays, the nighttime heat consumption is large, and the diurnal heat consumption is large on weekends, etc.

[0137] In view of the fact that the heat supply of the heat exchange unit corresponding to different building properties is different on weekdays and weekends, the target heat data of the heat exchange unit corresponding to different building properties at different time parameters (specifically, the target heat data of the heat exchange unit corresponding to different building properties at different time parameters is obtained from the obtained historical heat supply operation data) can be used to determine the variation law of the target heat data of the corresponding heat exchange unit at different time parameters, that is, the variation law of the target heat data of the heat exchange unit corresponding to each building property on weekdays and weekends is obtained.

[0138] On the basis of the above, if the target heat data of the heat exchange unit corresponding to the building property in the analysis period (a specific period in which the target heat data is analyzed to determine whether it is abnormal) exceeds the variation law of the target heat data of the heat exchange unit corresponding to the building property at the corresponding time parameter, it is determined that the target heat data in the analysis period is abnormal. For example, for an unattended office building on weekends, if the target heat data remains at the target heat data level on weekdays, it is determined to be abnormal, and vice versa. At this time, the normal target heat data under other working conditions can be used to repair the abnormal target heat data of the heat exchange unit corresponding to the building property at the corresponding time parameter (that is, the time parameter belonging to the same time parameter as the analysis period, for example, both are weekdays). The way of repairing the abnormal target heat data by using the normal target heat data under other working conditions is similar to the way of repairing the abnormal target heat data by using the normal target heat data under other working conditions when it is determined from the historical heat supply operation data that the target heat data and other heat data in the same working condition are abnormal, that is, the target heat data under other working conditions is used to repair the abnormal target heat data. The only difference is that the time parameter of the other working condition is the same as the time parameter of the analysis period.

[0139] By the above method, the building property and time parameter corresponding to the heat exchange unit are considered to further improve the reliability and accuracy of the target heat data cleaning.

[0140] The data cleaning method provided by the embodiment of the present application can further include:

[0141] Obtain the outdoor temperature corresponding to the historical heat supply operation data;

[0142] Determine the target heat data variation interval corresponding to the target heat data with a difference in outdoor temperature less than a preset temperature difference value;

[0143] If the target heat data corresponding to the outdoor temperature exceeds the corresponding target heat data variation interval, the normal target heat data under other working conditions is used to repair the abnormal target heat data.

[0144] In the present application, while the historical heat supply operation data of the heat exchange unit is acquired, the outdoor temperature (specifically, the average outdoor temperature) corresponding to the historical heat supply operation data is also acquired. Then, according to the target heat data corresponding to the outdoor temperature group in which the difference (for example, the difference between the highest average outdoor temperature and the lowest average outdoor temperature) of the outdoor temperature is less than the preset temperature difference value, the corresponding target heat data variation interval is determined by using a box plot or a 3σ principle. If the target heat data corresponding to the outdoor temperature exceeds the target heat data variation interval corresponding to the outdoor temperature, it is determined that the target heat data corresponding to the outdoor temperature is abnormal. That is, in the case where the outdoor temperature is relatively stable, if the target heat data in a certain period of time exceeds its reasonable variation interval, it is determined that the target heat data corresponding to the outdoor temperature is abnormal. For example: ① Human factors: the given heat value is not adjusted in conjunction with the meteorological parameters. ② Due to special reasons, the given heat is different from the reasonable value. At this time, the abnormal target heat data can be repaired by using the normal target heat data under other working conditions. The specific process is the same as the repair method described above, and will not be described here.

[0145] By the above-mentioned manner, the outdoor temperature factor is considered when cleaning the target heat data, so as to improve the reliability and accuracy of the target heat data cleaning.

[0146] Through the above process, it can be known that the present application applies many mechanisms to determine the target heat data, improves the quality of the target heat data cleaning, and at the same time of completing the data cleaning, better restores the heat demand of the heat user, provides guarantee for the heat user's heat characteristics tracked by the load prediction algorithm, so that the prediction effect is better.

[0147] The present application also provides a data cleaning device, which is shown in Figure 3 which shows a structure schematic diagram of a data cleaning device provided by the present application, which can include:

[0148] The first acquisition module 31 is configured to acquire the historical heat supply operation data of the heat exchange unit. The historical heat supply operation data can include heat data and other heat supply operation data other than heat data.

[0149] The first repair module 32 is configured to, if it is determined according to the historical heat supply operation data that the target heat data in the heat data under the same working condition is abnormal and other heat data is not all abnormal, repair the abnormal target heat data by using the other heat data under the same working condition which is not abnormal. The target heat data is the heat data participating in the load prediction.

[0150] The second repair module 33 is configured to repair the target heat data in the heat data under the same working condition if the target heat data and the other heat data in the heat data under the same working condition are determined to be abnormal according to the historical heat supply operation data.

[0151] The data cleaning device provided by the embodiment of the present application comprises a first determination module configured to determine that the target heat data in the heat data under the same working condition is abnormal and the other heat data is not all abnormal according to the historical heat supply operation data.

[0152] The first determination unit is configured to determine the reasonable variation interval of the target heat data and the reasonable variation interval of the other heat data by using a box plot or a 3σ principle.

[0153] The second determination unit is configured to determine that the target heat data in the heat data under the same working condition is abnormal and the other heat data is not all abnormal if the target heat data under the same working condition is greater than or smaller than the reasonable variation interval of the target heat data and the other heat data under the same working condition is within the reasonable variation interval of the corresponding other heat data.

[0154] The first repair module 32 can comprise:

[0155] The first repair unit is configured to repair the target heat data by using the other heat data with the best quality in the other heat data under the same working condition which is not abnormal.

[0156] The data cleaning device provided by the embodiment of the present application comprises a second determination module configured to determine that the target heat data and the other heat data in the heat data under the same working condition are all abnormal according to the historical heat supply operation data.

[0157] The third determination unit is configured to determine the reasonable variation interval of the primary network electric valve opening degree feedback by using a box plot or a 3σ principle when the heat regulating unit is operated in the time period adjustment mode and the electric valve is used for heat regulating, and determine that the corresponding target heat data and the other heat data are all abnormal if the primary network electric valve opening degree feedback is not within the reasonable variation interval of the primary network electric valve opening degree feedback during the time period adjustment or the primary network electric valve opening degree feedback is within the reasonable variation interval of the primary network electric valve opening degree feedback during the time period adjustment and is not within the reasonable variation sub-interval determined according to the reasonable variation interval of the primary network electric valve opening degree feedback.

[0158] The fourth determining unit is configured to determine a reasonable variation range of the secondary network temperature by using a box plot or a 3σ principle when the heat exchange unit operates in the time-division adjustment mode and the heat quantity is regulated by using the secondary network temperature adjustment, and determine that the corresponding target heat quantity data and other heat quantity data are abnormal if at least one of the secondary network temperatures is not in the reasonable variation range of the secondary network temperature during the time-division adjustment, or at least one of the secondary network temperatures is in the reasonable variation range of the secondary network temperature during the time-division adjustment and is not in a reasonable variation sub-range determined according to the reasonable variation range of the secondary network temperature; wherein the secondary network temperature can include a secondary network supply water temperature setting, a secondary network return water temperature setting, and a secondary network supply and return water temperature setting.

[0159] The second repair module 33 can include:

[0160] The processing unit is configured to process normal target heat quantity data in the same adjustment period as the abnormal target heat quantity data by using a box plot or a 3σ principle, and obtain a reasonable fluctuation range of the normal target heat quantity data.

[0161] The second repair unit is configured to fill in the abnormal target heat quantity data according to the reasonable fluctuation range of the normal target heat quantity data.

[0162] The data cleaning device provided by the embodiment of the present application can include a second determining module which can include:

[0163] The fifth determining unit is configured to determine reasonable variation ranges of the heat source instantaneous heat quantity, the heat source instantaneous flow quantity, the heat source supply water temperature, the primary network instantaneous heat quantity, and the primary network electric regulating valve opening degree feedback, respectively, by using a box plot or a 3σ principle; calculate a heat quantity transmission time length from the heat source to the analysis heat exchange unit by using the reasonable variation range of the heat source instantaneous heat quantity, the reasonable variation range of the heat source instantaneous flow quantity, the reasonable variation range of the heat source supply water temperature, and the reasonable variation range of the primary network instantaneous heat quantity; and determine that the corresponding target heat quantity data and other heat quantity data are abnormal if the fluctuation of the heat source instantaneous heat quantity exceeds a reasonable variation sub-range determined according to the reasonable variation range of the heat source instantaneous heat quantity, the fluctuation of the heat source instantaneous heat quantity reaches the analysis heat exchange unit after one heat quantity transmission time length, and the primary network electric regulating valve opening degree feedback value does not exceed its reasonable variation range.

[0164] The sixth determination unit is configured to determine reasonable variation intervals of the primary network instantaneous heat, the primary network instantaneous flow, the primary network water supply temperature, the primary network return water temperature and the primary network electric regulating valve opening degree respectively by using a box plot or a 3σ principle; calculate the heat transmission time length from the fluctuation heat exchange unit to the analysis heat exchange unit by using the reasonable variation interval of the primary network instantaneous heat, the reasonable variation interval of the primary network instantaneous flow, the reasonable variation interval of the primary network water supply temperature and the reasonable variation interval of the primary network return water temperature; if the fluctuation of the primary network instantaneous heat of the fluctuation heat exchange unit exceeds a reasonable variation subinterval determined according to the reasonable variation interval of the primary network instantaneous heat, the fluctuation of the primary network instantaneous heat of the heat exchange unit reaches the analysis heat exchange unit after one heat transmission time length, and the feedback of the primary network electric regulating valve opening degree does not exceed the reasonable variation interval thereof, it is determined that the corresponding target heat data and other heat data are all abnormal.

[0165] The seventh determination unit is configured to determine reasonable variation intervals of the primary network electric regulating valve opening degree given, the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat respectively by using a box plot or a 3σ principle; take a point where the primary network electric regulating valve opening degree given is not 0, the primary network electric regulating valve opening degree feedback and the primary network instantaneous heat are all 0 as a critical point according to the reasonable variation interval corresponding to the primary network electric regulating valve opening degree given, the reasonable variation interval corresponding to the primary network electric regulating valve opening degree feedback and the reasonable variation interval corresponding to the primary network instantaneous heat, and determine that the target heat data and other heat data below the critical point are all abnormal.

[0166] The eighth determination unit is configured to analyze the variation of the primary network instantaneous heat within a certain time span when the heat is regulated by using the secondary network temperature regulation, and obtain the fluctuation variation of the primary network instantaneous heat; when the peak amplitude and the valley amplitude in the fluctuation variation of the primary network instantaneous heat are relatively consistent, the time interval from the peak to the valley is relatively consistent, and the peak amplitude or the valley amplitude is greater than a preset amplitude or the time interval from the peak to the valley is abnormal, it is determined that the corresponding target heat data and other heat data are all abnormal.

[0167] The ninth determination unit is configured to determine that the corresponding target heat data and other heat data are all abnormal if at least one of the following conditions occurs: the equipment data is abnormal, the secondary network circulating water pump frequency feedback value is 0, the primary network electric regulating valve opening degree feedback is normal, the secondary network water supply temperature deviates from the reasonable variation interval thereof, and the primary network instantaneous heat is 0.

[0168] The tenth determination unit is configured to determine the reasonable variation range of the primary network electric regulating valve opening degree feedback by using a box plot or a 3σ principle; when the heat is regulated by switching from the primary network electric regulating valve opening degree feedback regulation to the secondary network temperature regulation, if the primary network electric regulating valve opening degree feedback is not within the reasonable variation range of the primary network electric regulating valve opening degree feedback, it is determined that the corresponding target heat data and other heat data are all abnormal.

[0169] The second repair module 33 can comprise:

[0170] A third repair unit is configured to, if the number of abnormal target heat data is less than a preset value, fill the abnormal target heat data with a mean value of normal target heat data of adjacent points.

[0171] A fourth repair unit is configured to, if the number of abnormal target heat data is not less than the preset value, fill the abnormal target heat data with normal target heat data within a preset time length before and a preset time length after.

[0172] The data cleaning device provided in the embodiment of the present application can comprise:

[0173] An eleventh determination unit is configured to determine whether the target heat data and other heat data in the heat data under the same working condition are missing.

[0174] A twelfth determination unit is configured to, if the target heat data and other heat data under the same working condition are all missing, determine that the target heat data and other heat data in the heat data under the same working condition are all abnormal.

[0175] The second repair module 33 can comprise:

[0176] A fifth repair unit is configured to, when the heat exchange unit is operated in a time period adjustment mode, process normal target heat data in the same adjustment time period as the abnormal target heat data by using a box chart or a 3σ principle to obtain a reasonable fluctuation range of the normal target heat data, and fill the abnormal target heat data according to the reasonable fluctuation range of the normal target heat data.

[0177] A sixth repair unit is configured to, when the heat exchange unit is operated in a time adjustment mode, if the number of abnormal target heat data is less than a preset value, fill the abnormal target heat data with a mean value of normal target heat data of adjacent points, and if the number of abnormal target heat data is not less than the preset value, fill the abnormal target heat data with normal target heat data within a preset time length before and a preset time length after.

[0178] The data cleaning device provided in the embodiment of the present application can comprise:

[0179] A thirteenth determination unit is configured to determine whether the target heat data and other heat data in the heat data under the same working condition are missing.

[0180] The fourteenth determination unit is configured to determine that the target heat data in the heat data under the same working condition is abnormal and other heat data is not all abnormal if the target heat data under the same working condition is missing and other heat data is not all missing.

[0181] The first repair module 32 can include:

[0182] The seventh repair unit is configured to repair the target heat data by using other heat data with optimal quality in other heat data under the same working condition which is not abnormal.

[0183] The data cleaning device provided in the embodiment of the application can further include:

[0184] The second acquisition module is configured to acquire time parameters corresponding to historical heat supply operation data and building properties corresponding to heat exchange units; the time parameters include weekdays and weekends.

[0185] The third determination module is configured to determine target heat data variation rules of the heat exchange units corresponding to different building properties at different time parameters according to target heat data of the heat exchange units corresponding to different building properties at different time parameters.

[0186] The third repair module is configured to repair abnormal target heat data by using normal target heat data of the heat exchange units corresponding to the building properties at corresponding time parameters and under other working conditions if the target heat data of the heat exchange units corresponding to the building properties within the analysis period exceeds the target heat data variation rules of the heat exchange units corresponding to the building properties at corresponding time parameters.

[0187] The data cleaning device provided in the embodiment of the application can further include:

[0188] The third acquisition module is configured to acquire outdoor temperatures corresponding to historical heat supply operation data.

[0189] The fourth determination module is configured to determine a target heat data variation interval according to target heat data corresponding to a difference value of the outdoor temperature less than a preset temperature difference value.

[0190] The fourth repair module is configured to repair abnormal target heat data by using normal target heat data under other working conditions if the target heat data corresponding to the outdoor temperature exceeds the target heat data variation interval.

[0191] The embodiment of the application further provides a data cleaning device, referring to Figure 4 which shows a structural schematic diagram of a data cleaning device provided in the embodiment of the application, and can include:

[0192] The memory 41 is configured to store a computer program.

[0193] The processor 42 is configured to execute the computer program stored in the memory 41, and implement the steps of any of the above data cleaning methods.

[0194] The data cleaning device and the related part of the data cleaning method provided in the embodiments of the present application can refer to the detailed description of the corresponding part of the data cleaning method provided in the embodiments of the present application, which will not be repeated here.

[0195] It should be noted that, in this document, the terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device comprising a series of elements inherently includes the elements. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, the above technical solutions provided in the embodiments of the present application are not described in detail, so as not to be too repetitive.

[0196] The above description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data cleaning method, characterized in that, include: Acquire historical heating operation data of the heat exchanger unit; the historical heating operation data includes heat data and other heating operation data besides the heat data; the heat data may specifically include the instantaneous heat of the primary network, the calculated heat of the primary network, the cumulative heat difference of the primary network, the instantaneous heat of the secondary network, the calculated heat of the secondary network, and the cumulative heat difference of the secondary network; other heating operation data may specifically include the instantaneous heat of the heat source, the instantaneous flow rate of the heat source, the supply water temperature of the heat source, the cumulative heat of the primary network, the supply water temperature of the primary network, the return water temperature of the primary network, the instantaneous flow rate of the primary network, the setpoint of the primary network electric regulating valve, the feedback of the opening of the primary network electric regulating valve, the manual / automatic switching of the primary network electric regulating valve, the cumulative heat of the secondary network, the supply water temperature of the secondary network, the return water temperature of the secondary network, the setpoint of the secondary network supply water temperature, the setpoint of the secondary network return water temperature, the setpoint of the secondary network supply and return water temperatures, the instantaneous flow rate of the secondary network, the setpoint of the secondary network circulating water pump frequency, the feedback of the secondary network circulating water pump frequency, the communication status, and the cumulative power consumption; If, based on the historical heating operation data, it is determined that the target heat data under the same operating condition is abnormal while other heat data are not all abnormal, then the abnormal target heat data is repaired using other heat data that are not abnormal under the same operating condition; the target heat data is the heat data participating in load forecasting; the same operating condition refers to the same point in time. If, based on the historical heating operation data, it is determined that both the target heat data and other heat data under the same operating condition are abnormal, then the abnormal target heat data will be repaired using the normal target heat data under other operating conditions. Specifically, determining whether the target heat data is abnormal and other heat data are not all abnormal based on the historical heating operation data includes: using a box plot or the 3σ principle to determine the reasonable variation range of the target heat data and the reasonable variation range of other heat data; if the target heat data under the same operating condition is greater than or less than the reasonable variation range of the target heat data, and there are other heat data under the same operating condition that are within the reasonable variation range of the corresponding other heat data, then it is determined that the target heat data under the same operating condition is abnormal and other heat data are not all abnormal. The method of repairing abnormal target heat data using other heat data that are not abnormal under the same operating condition includes: repairing the target heat data using the best quality other heat data among the other heat data that are not abnormal under the same operating condition; the method of determining the best quality other heat data includes: using historical heating operation data within a preset historical period to cross-verify data groups composed of other heat data of the same type under different operating conditions, and determining the target data group with the highest percentage of non-abnormal other heat data based on the obtained verification results, and determining the other heat data in the target data group that belong to the same operating condition as the abnormal target heat data as the best quality other heat data; or, determining the best quality other heat data based on the average of other heat data belonging to the same operating condition as the abnormal target heat data in two or more groups of non-abnormal other heat data.

2. The data cleaning method according to claim 1, characterized in that, Based on the historical heating operation data, it was determined that the target heat data and other heat data under the same operating condition were abnormal, including: When the heat exchanger unit operates in time-sharing regulation mode and uses electric regulating valves for heat control, the reasonable variation range of the primary grid electric regulating valve opening feedback is determined by box plot or 3σ principle. If the primary grid electric regulating valve opening feedback is not within the reasonable variation range of the primary grid electric regulating valve opening feedback during the time-sharing regulation period, or if the primary grid electric regulating valve opening feedback is within the reasonable variation range of the primary grid electric regulating valve opening feedback during the time-sharing regulation period but not within the reasonable variation sub-interval determined based on the reasonable variation range of the primary grid electric regulating valve opening feedback, then the corresponding target heat data and other heat data are determined to be abnormal. When the heat exchanger unit operates in time-sharing regulation mode and uses secondary network temperature regulation for heat control, the reasonable variation range of the secondary network temperature is determined using a box plot or the 3σ principle. If at least one of the secondary network temperatures is not within the reasonable variation range during the time-sharing regulation period, or if at least one of the secondary network temperatures is within the reasonable variation range during the time-sharing regulation period but not within the reasonable sub-range determined based on the reasonable variation range of the secondary network temperature, then the corresponding target heat data and other heat data are determined to be abnormal. The secondary network temperature includes the secondary network supply water temperature setting, the secondary network return water temperature setting, and the secondary network supply and return water temperature settings. Repairing abnormal target heat data using normal target heat data under other operating conditions includes: By using box plots or the 3σ principle to process normal target heat data that are in the same adjustment period as abnormal target heat data, a reasonable fluctuation range for normal target heat data can be obtained. Abnormal target heat data are filled in based on the reasonable fluctuation range of normal target heat data.

3. The data cleaning method according to claim 1, characterized in that, Based on the historical heating operation data, it was determined that the target heat data and other heat data under the same operating condition were abnormal, including: The reasonable variation ranges corresponding to the instantaneous heat source heat, instantaneous flow rate, heat source water supply temperature, primary network instantaneous heat, and primary network electric regulating valve opening feedback are determined using box plots or the 3σ principle. The heat transfer time from the heat source to the analytical heat exchanger is calculated using these reasonable variation ranges. If the fluctuation of the instantaneous heat source heat exceeds the reasonable variation sub-range determined based on the reasonable variation range of the instantaneous heat source heat, and the fluctuation reaches the analytical heat exchanger after one heat transfer time, and the primary network electric regulating valve opening feedback value does not exceed its reasonable variation range, then the corresponding target heat data and other heat data are determined to be abnormal. Using box plots or the 3σ principle, reasonable variation ranges are determined for the instantaneous heat, instantaneous flow, supply water temperature, return water temperature, and opening of the primary network's electric regulating valve. Using these reasonable variation ranges for the primary network's instantaneous heat, flow, supply water temperature, and return water temperature, the heat transfer time from the fluctuating heat exchanger unit to the analytical heat exchanger unit is calculated. If the fluctuation of the instantaneous heat in the primary network of the fluctuating heat exchanger unit exceeds the reasonable variation sub-range determined based on the reasonable variation range of the primary network's instantaneous heat, and the fluctuation of the instantaneous heat in the primary network of the heat exchanger unit reaches the analytical heat exchanger unit after one heat transfer time, and the feedback of the primary network's electric regulating valve opening does not exceed its reasonable variation range, then the corresponding target heat data and other heat data are determined to be abnormal. Using box plots or the 3σ principle, reasonable variation ranges are determined for the primary grid electric control valve opening setpoint, the primary grid electric control valve opening feedback, and the primary grid instantaneous heat. Based on the reasonable variation ranges corresponding to the primary grid electric control valve opening setpoint, the primary grid electric control valve opening feedback, and the primary grid instantaneous heat, the point where the primary grid electric control valve opening setpoint is not 0, the primary grid electric control valve opening feedback, and the primary grid instantaneous heat all become 0 are taken as critical points. It is determined that the target heat data and other heat data below the critical point are all abnormal. When using secondary network temperature regulation for heat control, the changes in instantaneous heat of the primary network over a certain time span are analyzed to obtain the fluctuation changes of instantaneous heat of the primary network. When the peak amplitude and trough amplitude of the fluctuation changes of instantaneous heat of the primary network are relatively consistent, and the time interval between the peak and the trough is relatively consistent, if there are peak amplitudes or trough amplitudes that are greater than the preset amplitude or if the time interval between the peak and the trough is abnormal, then it is determined that the corresponding target heat data and other heat data are abnormal. If any of the following situations occur: abnormal equipment data, 0 frequency feedback value of secondary network circulating water pump and normal opening feedback of primary network electric regulating valve and secondary network water supply temperature deviating from its reasonable variation range, or 0 instantaneous heat of primary network, then it is determined that the corresponding target heat data and other heat data are abnormal. The reasonable range of variation of the primary network electric regulating valve opening feedback is determined by using a box plot or the 3σ principle. When switching from heat control using the primary network electric regulating valve opening feedback to heat control using the secondary network temperature regulation, if the primary network electric regulating valve opening feedback is not within the reasonable range of variation of the primary network electric regulating valve opening feedback, then the corresponding target heat data and other heat data are determined to be abnormal. Repairing abnormal target heat data using normal target heat data under other operating conditions includes: If the number of abnormal target heat data is less than the preset value, the abnormal target heat data will be filled by the average of the normal target heat data of the adjacent points. If the number of abnormal target heat data is not less than the preset value, the abnormal target heat data will be filled in using normal target heat data within the previous preset time period and normal target heat data within the subsequent preset time period.

4. The data cleaning method according to claim 1, characterized in that, Based on the historical heating operation data, it was determined that the target heat data and other heat data under the same operating condition were abnormal, including: Determine whether there are any missing target heat data and other heat data in the heat data under the same operating condition; If all target heat data and other heat data under the same operating condition are missing, then it is determined that the target heat data and other heat data under the same operating condition are abnormal. Repairing abnormal target heat data using normal target heat data under other operating conditions includes: When the heat exchanger unit is operating in time-sharing regulation mode, the box plot or 3σ principle is used to process the normal target heat data that is in the same regulation period as the abnormal target heat data to obtain the reasonable fluctuation range of the normal target heat data; and the abnormal target heat data is filled in according to the reasonable fluctuation range of the normal target heat data. When the heat exchanger unit is operating in hourly adjustment mode, if the number of abnormal target heat data is less than the preset value, the abnormal target heat data is filled by the average of the normal target heat data of adjacent points; if the number of abnormal target heat data is not less than the preset value, the abnormal target heat data is filled by the normal target heat data within the previous preset time length and the normal target heat data within the subsequent preset time length.

5. The data cleaning method according to claim 1, characterized in that, Based on the historical heating operation data, it is determined that the target heat data is abnormal and other heat data are not all abnormal under the same operating condition, including: Determine whether there are any missing target heat data and other heat data in the heat data under the same operating condition; If the target heat data under the same operating condition is missing, but other heat data is not all missing, then it is determined that the target heat data under the same operating condition is abnormal, but other heat data is not all abnormal. Repairing abnormal target heat data using other heat data that are not abnormal under the same operating conditions, including: The target heat data is repaired by using the best quality heat data from other heat data that are not abnormal under the same operating conditions.

6. The data cleaning method according to claim 1, characterized in that, Also includes: Obtain the time parameters corresponding to the historical heating operation data and the building properties corresponding to the heat exchange unit; The time parameters are divided into weekdays and rest days; Based on the target heat data of heat exchange units corresponding to different building types at different time parameters, determine the variation law of the target heat data of the corresponding heat exchange units at different time parameters; If the target heat data of the heat exchange unit of the corresponding building type exceeds the target heat data change pattern of the corresponding building type heat exchange unit in the corresponding time parameter, then the abnormal target heat data is repaired by using the normal target heat data of the heat exchange unit of the corresponding building type in the corresponding time parameter and under other operating conditions.

7. The data cleaning method according to claim 1, characterized in that, Also includes: Obtain the outdoor temperature corresponding to the historical heating operation data; The target heat data variation range is determined based on the target heat data corresponding to the outdoor temperature difference being less than the preset temperature difference value; If the target heat data corresponding to the outdoor temperature exceeds the corresponding target heat data variation range, the abnormal target heat data will be repaired using normal target heat data under other operating conditions.

8. A data cleaning apparatus, characterized in that, include: The first acquisition module is used to acquire historical heating operation data of the heat exchanger unit. The historical heating operation data includes heat data and other heating operation data besides the heat data. The heat data may specifically include the instantaneous heat of the primary network, the calculated heat of the primary network, the cumulative heat difference of the primary network, the instantaneous heat of the secondary network, the calculated heat of the secondary network, and the cumulative heat difference of the secondary network. Other heating operation data may specifically include the instantaneous heat of the heat source, the instantaneous flow rate of the heat source, the supply water temperature of the heat source, the cumulative heat of the primary network, the supply water temperature of the primary network, the return water temperature of the primary network, the instantaneous flow rate of the primary network, the set opening degree of the primary network electric regulating valve, the feedback opening degree of the primary network electric regulating valve, the manual / automatic switching of the primary network electric regulating valve, the cumulative heat of the secondary network, the supply water temperature of the secondary network, the return water temperature of the secondary network, the set supply water temperature of the secondary network, the set return water temperature of the secondary network, the set supply and return water temperature of the secondary network, the instantaneous flow rate of the secondary network, the set frequency of the secondary network circulating water pump, the feedback frequency of the secondary network circulating water pump, the communication status, and the cumulative power consumption. The first repair module is used to repair the abnormal target heat data by using other heat data that are not abnormal under the same operating condition if the target heat data is abnormal and other heat data are not all abnormal, based on the historical heating operation data; the target heat data is the heat data that participates in load forecasting; the same operating condition is the same point in time. The second repair module is used to repair the abnormal target heat data by using normal target heat data under other operating conditions if it is determined from the historical heating operation data that the target heat data and other heat data under the same operating condition are abnormal. Specifically, the first repair module is used to determine the reasonable variation range of the target heat data and the reasonable variation range of other heat data using a box plot or the 3σ principle; if the target heat data under the same working condition is greater than or less than the reasonable variation range of the target heat data, and there are other heat data under the same working condition that are within the reasonable variation range of the corresponding other heat data, then it is determined that the target heat data under the same working condition is abnormal and the other heat data are not all abnormal. Specifically, the first repair module is used to repair the target heat data by using the highest quality heat data among other heat data that are not abnormal under the same operating condition. The method for determining the highest quality heat data includes: using historical heating operation data within a preset historical period to mutually verify data groups composed of other heat data of the same type under different operating conditions, and determining the target data group with the highest percentage of non-abnormal heat data based on the obtained verification results, and determining the other heat data in the target data group that belong to the same operating condition as the abnormal target heat data as the highest quality heat data; or, determining the highest quality heat data based on the average of the other heat data belonging to the same operating condition as the abnormal target heat data in two or more groups of non-abnormal heat data.

9. A data cleaning device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the data cleaning method as described in any one of claims 1 to 7.

Images