A continuous optimization heating load prediction method and system based on big data analysis learning

Abstract

The present application relates to the field of heating heat load operation regulation, and provides a continuous optimization heating load prediction method and system based on big data analysis learning, comprising: step S1: configuring a data interface on a real-time library service module to receive real-time and historical operation data, the operation data including meteorological information data, heat load of a heat exchange station, and room temperature qualification rate; step S2: preprocessing the collected operation data through a data processing module integrated on a heating platform service module, and analyzing the preprocessed operation data through an analysis algorithm stored in a relationship module to determine a predicted heat load value of the next control cycle. By combining a heating platform, learning and analyzing historical heat network operation data using big data analysis technology, and making appropriate corrections and adjustments, the heat load of the heating system can be effectively predicted, which is closely combined with actual operation, significantly improves prediction accuracy, simplifies operation process and optimizes energy use, and improves operation efficiency and system response speed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of heating heat load operation regulation, and particularly relates to a continuous optimization heating load prediction method and system based on big data analysis and learning. BACKGROUND

[0002] Heating system heat load prediction is an important part of heating system planning and the basis for its economic and efficient operation. Accurate prediction of heat load can provide strong data support for the operation and management of the heating system, thereby reasonably guiding the operation and regulation management work. Current prediction techniques mainly rely on complex mathematical models and formulas. Although these methods are theoretically feasible, in practical applications, due to the complexity of the model and the delay in response to external variables, they often result in inaccurate predictions or limited application range. Therefore, it is particularly important to develop a prediction method that is both accurate and easy to implement. SUMMARY

[0003] To solve the above problems, the present application provides a continuous optimization heating load prediction method and system based on big data analysis and learning, which realizes dynamic prediction and real-time adjustment of heating load by comprehensively utilizing historical and real-time meteorological data, heating system performance data and user demand changes, and solves the problems of high complexity, insufficient prediction accuracy and poor practicability in existing heat load prediction methods. The above invention purpose of the present application is realized by the following technical scheme:

[0004] The present application provides a continuous optimization heating load prediction method based on big data analysis and learning, comprising,

[0005] Step S1: configuring a data interface on a real-time library service module to receive real-time and historical operation data, the operation data including meteorological information data, heat load of heat exchange station and room temperature qualification rate;

[0006] Step S2: preprocessing the collected operation data through a data processing module integrated on a heating platform service module, and analyzing the preprocessed operation data through an analysis algorithm stored in a relationship module to determine the predicted heat load value of the next control cycle;

[0007] Wherein, the analysis algorithm stored in the relationship module is used to analyze the preprocessed operation data to determine the predicted heat load value of the next control cycle, including, based on the real-time collected operation data, the corrected comprehensive outdoor temperature under different meteorological information, heat load of heat exchange station and room temperature qualification rate are counted; based on different meteorological information and comprehensive outdoor temperature, the next control cycle is searched and positioned according to weather category, and the similar meteorological information condition is found by using the minimum comprehensive outdoor temperature difference value principle to obtain the corresponding predicted heat load value.

[0008] Further, in step S2, the operation data is pre-processed by a data processing module integrated on the heat supply platform service module, including validity processing of the operation data, including removing operation data within a preset time after heat network operation adjustment, and standardizing, detecting and cleaning the operation data.

[0009] Further, in step S2, based on the real-time collected operation data, the corrected converted comprehensive outdoor temperature, heat load of the heat exchange station and room temperature qualified rate under different meteorological information are counted, including real-time counting of the heat load of the heat exchange station and the room temperature qualified rate under different meteorological information, and calculating the converted comprehensive outdoor temperature by the conversion correction algorithm, wherein the meteorological information includes weather type, temperature and wind speed.

[0010] Further, in step S2, the conversion correction algorithm is,

[0011] When the current wind speed is lower than the baseline wind speed, the converted comprehensive outdoor temperature T is the actual outdoor temperature t;

[0012] When the current wind speed is between the baseline wind speed and the incremental wind speed threshold, the calculation formula of the converted comprehensive outdoor temperature T is, ; wherein, is the baseline wind speed, and v is the current wind speed;

[0013] When the current wind speed reaches the incremental wind speed threshold but does not exceed the medium wind speed threshold, the calculation formula of the converted comprehensive outdoor temperature T is, , wherein, is the incremental wind speed threshold, b is the basic adjustment value, and c and d are related adjustment coefficients of wind speed increment;

[0014] When the current wind speed exceeds the medium wind speed threshold to reach high wind speed, the calculation formula of the converted comprehensive outdoor temperature T is, , and k is a high wind speed preset fixed reduction value.

[0015] Further, in step S2, based on different meteorological information and converted comprehensive outdoor temperature, the next control cycle searches and locates the weather category according to the minimum converted comprehensive outdoor temperature difference value principle to find similar meteorological information conditions to obtain the corresponding predicted heat load value, including,

[0016] An expected value of the room temperature qualified rate is set to evaluate whether the heat supply effect of the current heat exchange station reaches the expectation;

[0017] The room temperature qualified rate under the same meteorological information condition is judged;

[0018] When the room temperature qualified rate reaches or exceeds the expected value, the predicted heat load is set as the historical heat exchange station heat load under similar meteorological information conditions;

[0019] When the room temperature qualification rate does not reach the expected value, the predicted heat load is set as the heat load of the heat exchange station under similar meteorological information conditions plus an adjustment value β, and the adjustment value β is the heat load amount used for adjustment when the room temperature qualification rate exceeds the expected value.

[0020] Further, the similar meteorological information condition is the heat load of the historical heat exchange station with the minimum difference in the converted comprehensive outdoor temperature under the same weather type.

[0021] Further, the method further comprises step S3: sending the predicted heat load value to the heating control system as the basis for heating load adjustment; and the heating control system adjusts the operating parameters of the heat exchange station according to the received predicted heat load value and feeds back the operating data to the real-time library service module.

[0022] Based on the same inventive concept, the present application also provides a continuous optimization heating load prediction system based on big data analysis and learning, which adopts the continuous optimization heating load prediction method as described above, and comprises,

[0023] The real-time library service module is configured to configure a data interface to receive real-time and historical operating data, and the operating data includes meteorological information data, heat load of the heat exchange station, and room temperature qualification rate.

[0024] The relationship module is configured to store an analysis algorithm for analyzing the preprocessed operating data.

[0025] The data processing module is configured to preprocess the collected operating data and analyze the preprocessed operating data by using the analysis algorithm stored in the relationship module to determine the predicted heat load value of the next control cycle. The analysis of the preprocessed operating data by using the analysis algorithm stored in the relationship module to determine the predicted heat load value of the next control cycle comprises: based on the real-time collected operating data, the corrected converted comprehensive outdoor temperature, the heat load of the heat exchange station, and the room temperature qualification rate under different meteorological information are counted; based on the different meteorological information and the converted comprehensive outdoor temperature, the weather category is located according to the search in the next control cycle, and the corresponding predicted heat load value is found under the similar meteorological information condition by using the minimum converted comprehensive outdoor temperature difference principle.

[0026] Further, the data processing module comprises,

[0027] The data preprocessing unit is configured to perform validity processing on the operating data, including removing the operating data within a preset time after the heat network operation adjustment, and performing standardization, abnormal value detection, and cleaning on the operating data.

[0028] The data statistical unit is used for collecting the running data in real time, and is used for statistically calculating the corrected converted comprehensive outdoor temperature, the heat load of the heat exchange station and the room temperature qualified rate under different weather information, further comprising that the heat load of the heat exchange station and the room temperature qualified rate under different weather information are statistically calculated in real time, and the converted comprehensive outdoor temperature is calculated through the conversion correction algorithm, wherein the weather information comprises weather type, temperature and wind speed.

[0029] The data analysis unit is used for setting an expected value of the room temperature qualified rate, and is used for evaluating whether the heating effect of the current heat exchange station reaches the expectation; the room temperature qualified rate under the same weather information condition is judged; when the room temperature qualified rate reaches or exceeds the expected value, the predicted heat load is set as the historical heat load of the heat exchange station under the similar weather information condition; when the room temperature qualified rate does not reach the expected value, the predicted heat load is set as the heat load of the heat exchange station under the similar weather information condition plus an adjustment value β, and the adjustment value β is the heat load amount used for adjustment when the room temperature qualified rate exceeds the expected value.

[0030] Further, the data feedback module is further used for sending the predicted heat load value to the heating control system as the basis for adjusting the heating load; meanwhile, the heating control system adjusts the operation parameters of the heat exchange station according to the received predicted heat load value and feeds back the operation data to the real-time database service module.

[0031] Compared with the prior art, the present application has at least one of the following beneficial effects:

[0032] The present application combines the heating platform, learns and analyzes the historical heat network operation data by using the big data analysis technology, and appropriately corrects and adjusts, so as to effectively predict the heat load of the heating system. The method is novel and simple in thought, practical, significantly improves the prediction accuracy, simplifies the operation process and optimizes the energy use, and improves the operation efficiency and system response speed. (1) The big data analysis learning method is used to predict the heating load. Compared with the traditional mathematical model or statistical method, the method can more accurately process and analyze complex data, and improve the prediction accuracy. The method can better adapt to the dynamic changes of the environment and user demand. (2) Real-time prediction of the heating load is realized, which can quickly respond to the changes of external weather and internal load demand, timely adjust the heating parameters, and ensure the efficient operation of the heating system. At the same time, accurate load prediction helps to reasonably arrange the output of the heat source, avoid the problems of excessive heating or insufficient heating, and reduce energy consumption. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 The step flow chart of the present application based on the continuous optimization of the heating load prediction method of the big data analysis learning;

[0034] Figure 2System schematic diagram of the continuous optimization of heating load prediction system based on big data analysis learning of the present application;

[0035] Figure 3 Heat load and room temperature qualified rate statistical table of the present application under different meteorological information;

[0036] Figure 4 Data statistical analysis learning rule table of the present application. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme 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 part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0038] Those skilled in the art can understand that, unless specifically stated, the singular forms "a", "an" and "the" used herein also include the plural forms. It should be further understood that the phrase "comprising" used in the specification of the present application means that the features, integers, steps, operations, elements and / or components exist, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0039] First embodiment

[0040] In modern heating systems, heating load prediction is one of the key technologies to ensure the economic and efficient operation of the system. Accurate heating load prediction can help operators optimize heating plans and adjustment strategies, thereby minimizing energy consumption and maximizing user comfort. However, the current heating load prediction methods mainly based on traditional mathematical modeling techniques, including but not limited to linear regression models, time series analysis and complex artificial neural networks, etc. These methods, while providing solutions to a certain extent, also have many limitations. Specifically, the existing heating system heating load prediction technology mainly relies on complex mathematical models, which not only increases the operation difficulty and limits the use of non-professionals, but also due to the lack of universality and adaptability of the model, it often cannot accurately adapt to rapidly changing weather conditions and user demands, resulting in low prediction accuracy. In addition, these technologies fail to effectively utilize the large amount of real-time data generated, so that the potential value of data resources is not fully tapped, thereby affecting the practicality and market adaptability of the technology. Therefore, it is necessary to develop a new prediction method to solve these problems of existing methods and provide a more efficient, accurate and user-friendly heating load prediction solution.

[0041] Based on this, the inventors, based on in-depth thinking and analysis of the defects of the prior art, thought of solving these problems by combining a heat supply informationization operation and management platform and using big data analysis technology. This method can make full use of historical and real-time data, learn patterns in the data and make appropriate corrections and adjustments, not only simplifying the operation process, improving the real-time and accuracy of the prediction, but also enhancing the adaptability and practicality of the system, thereby more effectively supporting the economic and efficient operation of the heat supply system. The specific implementation is as follows:

[0042] As shown in Figure 1 The present application provides a continuous optimization of heat load prediction method based on big data analysis learning, comprising,

[0043] Step S1: configuring a data interface on a real-time library service module to receive real-time and historical operation data, i.e. historical heat supply system data sources, the operation data including meteorological information data, heat load of heat exchange station and room temperature qualification rate;

[0044] Step S2: pre-processing the collected operation data through the data processing module integrated on the heat supply platform service module, and analyzing the pre-processed operation data through the analysis algorithm stored in the relationship module to determine the predicted heat load value of the next control cycle;

[0045] Among them, the analysis algorithm stored in the relationship module is used to analyze the pre-processed operation data to determine the predicted heat load value of the next control cycle, including based on the real-time collected operation data, counting the corrected comprehensive outdoor temperature, heat load of heat exchange station and room temperature qualification rate under different meteorological information; based on different meteorological information and comprehensive outdoor temperature, searching and positioning the weather category according to the next control cycle, and finding similar meteorological information conditions by using the principle of minimum comprehensive outdoor temperature difference to obtain the corresponding predicted heat load value.

[0046] Among them, in step S2, the data processing module integrated on the heat supply platform service module is used to pre-process the operation data, including validity processing of the operation data, including removing the operation data within a preset time after heat network operation adjustment, and standardizing, detecting and cleaning the operation data.

[0047] Specifically, the data processing module integrated on the heat supply platform service module is used to realize the validity processing of the operation data source. Since the heat supply system has hysteresis, the data within 2 hours after the heat load adjustment of the heat network operation is removed to ensure that the heat network has stabilized. The specific operation mode includes time window filtering of the data, which removes the data within 2 hours after adjustment by setting a specific time window, ensuring that the data used for analysis and prediction is more representative and stable.

[0048] Meanwhile, in step S2, based on the real-time collected operation data, the corrected converted comprehensive outdoor temperature, heat load of the heat exchange station and room temperature qualified rate under different meteorological information are counted, including real-time counting of the heat load of the heat exchange station and the room temperature qualified rate under different meteorological information, and calculating the converted comprehensive outdoor temperature through the conversion correction algorithm, wherein the meteorological information includes weather type, temperature and wind speed, the converted comprehensive outdoor temperature is to convert the wind force, an important influencing factor of the heat load, into the outdoor temperature through certain reasonable rules, and the conversion correction algorithm is,

[0049] ① When the current wind speed is lower than the baseline wind speed, the converted comprehensive outdoor temperature T is the actual outdoor temperature t;

[0050] ② When the current wind speed is between the baseline wind speed and the incremental wind speed threshold, the calculation formula of the converted comprehensive outdoor temperature T is, ; wherein, is the baseline wind speed, and v is the current wind speed;

[0051] ③ When the current wind speed reaches the incremental wind speed threshold but does not exceed the medium wind speed threshold, the calculation formula of the converted comprehensive outdoor temperature T is, , wherein, is the incremental wind speed threshold, b is the basic adjustment value, and c and d are related adjustment coefficients of the wind speed increment;

[0052] ④ When the current wind speed exceeds the medium wind speed threshold to reach the high wind speed, the calculation formula of the converted comprehensive outdoor temperature T is, , and k is a high wind speed preset fixed reduction value.

[0053] Specifically, as shown in Figure 3 , the heat load of the heat exchange station and the room temperature qualified rate under different meteorological information are counted, including weather such as sunny day, cloudy day, rainy day and snowy day, room temperature t, wind force, i.e. wind speed v, converted comprehensive outdoor temperature T, heat load Q of the heat exchange station and room temperature qualified rate G; according to the above correction algorithm,

[0054]

[0055] , wherein t is the outdoor temperature, v is the wind speed, and T is the converted comprehensive outdoor temperature.

[0056] As shown in the formula, when the wind speed , in this baseline wind speed range, the influence of wind on the feeling temperature is small, so there is no adjustment to the original temperature t; when , in this incremental wind speed range, with the increase of the wind speed, the feeling temperature is moderately affected. Therefore, v-2 in the formula represents the direct linear negative influence of the wind speed exceeding 2 m / s on the temperature; when the wind speed In this wind speed range, the influence of wind speed on the felt temperature is more significant, so the adjustment formula includes not only a basic adjustment value 3, but also an increasing factor , namely , where v-5 represents the part exceeding 5 m / s, and 0.2 is an empirical coefficient representing the trend that the influence of wind speed increase on temperature is enhanced; when the wind speed is high, a fixed reduction value for high wind speed is directly subtracted from the temperature.

[0057] Further, in step S2, based on different meteorological information, the comprehensive outdoor temperature is converted, and in the next control cycle, the weather category is searched and located, and the minimum difference principle of the converted comprehensive outdoor temperature is used to find the similar meteorological information condition to obtain the corresponding predicted heat load value, including

[0058] An expected value of the room temperature qualified rate is set to evaluate whether the heating effect of the current heat exchange station reaches the expectation;

[0059] The room temperature qualified rate under the same meteorological information condition is used for judgment;

[0060] When the room temperature qualified rate reaches or exceeds the expected value, the predicted heat load is set as the historical heat exchange station heat load under the similar meteorological information condition; when the room temperature qualified rate does not reach the expected value, the predicted heat load is set as the heat exchange station heat load under the similar meteorological information condition plus an adjustment value β, which is the heat load amount used for adjustment when the room temperature qualified rate exceeds the expected value. Among them, the similar meteorological information condition is the heat load of the historical heat exchange station with the minimum difference of the converted comprehensive outdoor temperature under the same weather type

[0061] Specifically, as Figure 4As shown, if the room temperature qualification rate G is greater than or equal to 95%, the predicted heat load under the same condition in the next cycle is assigned Q; if G is less than 95%, the predicted heat load under the same condition in the next cycle is assigned Q+β, β is an adjustment value, which is determined by the operating personnel according to the room temperature qualification rate and experience, and the setting range of β is 1%-5% of the heat load, and considering factors such as the granularity, accuracy of weather prediction, heat after nature of the heat exchange station pipe network, different periods of multiple learning rules of 1h, 2h and 4h are set, and learning knowledge bases corresponding to 0.5h, 1h, 2h and 4h are established. And the learning prediction is obtained according to the weather and the converted comprehensive outdoor temperature in the next cycle, the data statistical learning rule table is searched, and the predicted heat load is obtained. At the same time, based on the predicted load obtained by the learning algorithm of the heating operation management platform, the predicted data is sent to the control system of the heating control system (DCS / PLC) to provide the basis for adjusting the output of the heat exchange station, and the running data is collected and obtained in real time, and the learning algorithm knowledge base is constantly enriched and improved, which provides guarantee for improving the prediction and providing accurate optimal operation load. The predicted heat load running effect of the learning algorithm under different periods is compared and analyzed, the analysis learning prediction algorithm under the condition of meeting the environmental object attribute is obtained, and the knowledge base is continuously learned and improved, and the control optimization operation of the heat exchange station is guided.

[0062] Further, it further comprises the step S3 of sending the predicted heat load value to the heating control system as the basis for heating load adjustment; at the same time, the heating control system adjusts the operation parameters of the heat exchange station according to the received predicted heat load value and feeds back the operation data to the real-time library service module.

[0063] In summary, the heating load prediction method based on big data analysis learning proposed by the present application shows significant advantages and improvements compared with the prior art. First of all, this method integrates the rich data resources collected by the heating platform system, uses advanced big data analysis technology, realizes dynamic prediction and real-time adjustment of heat load, simplifies the operation process, and improves the practicability and applicability of the method. Secondly, the present application uses the room temperature qualification rate as the basis for heat load correction, which enhances the reliability of the prediction, and allows the system to make online correction according to real-time data, so as to more accurately adapt to the actual operating conditions. Finally, with the accumulation and analysis of more and more data, the present application can continuously optimize the prediction model, improve the accuracy of heat load prediction, and ensure the efficiency and economy of the heating system operation. Therefore, the present application not only has advanced technology, but also conforms to the trend of intelligentization and data-driven management of modern heating systems, and has wide application prospect and practical value.

[0064] Second embodiment

[0065] Based on the same inventive concept, as Figure 2As shown, the application also proposes a continuous optimization heating load prediction system based on big data analysis learning, which realizes a hardware system composition of: a heating platform, a real-time module, and a relationship module. The real-time database module is used for real-time data storage and connection with historical data sources, including meteorological information data, heat load data, and room temperature qualification rate data. The relationship database module is used for storing parameter relationship information and statistical learning tables. The heating platform is the core of the big data analysis learning prediction method, and has built-in data processing, analysis, and load prediction rule method programs. The continuous optimization heating load prediction method as described above includes,

[0066] A real-time database service module is configured to receive real-time and historical operation data, and the operation data includes meteorological information data, heat load of a heat exchange station, and room temperature qualification rate.

[0067] A relationship module is configured to store analysis algorithms for analyzing the preprocessed operation data.

[0068] A data processing module is configured to preprocess the collected operation data and analyze the preprocessed operation data through the analysis algorithms stored in the relationship module to determine the predicted heat load value of the next control cycle. The analysis of the preprocessed operation data through the analysis algorithms stored in the relationship module to determine the predicted heat load value of the next control cycle includes: based on the real-time collected operation data, the corrected converted comprehensive outdoor temperature, the heat load of the heat exchange station, and the room temperature qualification rate under different meteorological information are counted; based on different meteorological information and converted comprehensive outdoor temperature, the next control cycle is searched and positioned according to the weather category, and the corresponding predicted heat load value is found under the condition of similar meteorological information by using the principle of minimum converted comprehensive outdoor temperature difference.

[0069] Further, the data processing module includes,

[0070] A data preprocessing unit is configured to perform validity processing on the operation data, including removing the operation data within a preset time after heat network operation adjustment, and performing standardization, anomaly detection, and cleaning on the operation data.

[0071] A data statistical unit is configured to count the corrected converted comprehensive outdoor temperature, the heat load of the heat exchange station, and the room temperature qualification rate under different meteorological information based on the real-time collected operation data, and further includes: the heat load of the heat exchange station and the room temperature qualification rate under different meteorological information are counted in real time, and the converted comprehensive outdoor temperature is calculated through a conversion correction algorithm, wherein the meteorological information includes weather type, temperature, and wind speed.

[0072] The data analysis unit is configured to set an expected value of a room temperature qualification rate for evaluating whether the heat supply effect of the current heat exchange station reaches the expectation, to judge the room temperature qualification rate under the same meteorological information condition, to set the predicted heat load as the historical heat exchange station heat load under the similar meteorological information condition when the room temperature qualification rate reaches or exceeds the expected value, and to set the predicted heat load as the heat exchange station heat load under the similar meteorological information condition plus an adjustment value β when the room temperature qualification rate does not reach the expected value, the adjustment value β being the heat load amount for adjustment when the room temperature qualification rate exceeds the expected value.

[0073] Further, the data feedback module is configured to send the predicted heat load value to the heat supply control system as the basis for heat supply load adjustment, and the heat supply control system adjusts the operation parameters of the heat exchange station according to the received predicted heat load value and feeds back the operation data to the real-time database service module.

[0074] The above description is only the preferred embodiments of the present application, and the protection scope of the present application is not limited to the above embodiments. Any technical solution falling within the concept of the present application shall fall within the protection scope of the present application. It should be noted that, for ordinary skilled persons in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements shall also be considered as falling within the protection scope of the present application.

[0075] It should be noted that the above embodiments can be freely combined as needed. The above description is only the preferred embodiments of the present application, and it should be noted that, for ordinary skilled persons in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements shall also be considered as falling within the protection scope of the present application.

Claims

1. A method for continuous optimization of heating load forecasting based on big data analytics and learning, characterized in that, include, Step S1: Configure the data interface on the real-time library service module to receive real-time and historical operating data, including meteorological information data, heat load of heat exchange station and room temperature qualification rate; Step S2: The collected operating data is preprocessed by the data processing module integrated on the heating platform service module, and the preprocessed operating data is analyzed by the analysis algorithm stored in the relationship module to determine the predicted heat load value for the next control cycle. The analysis of the preprocessed operating data to determine the predicted heat load value for the next control cycle by the analysis algorithm stored in the relationship module includes: based on the real-time collected operating data, statistically analyzing the corrected comprehensive outdoor temperature, the heat load of the heat exchange station, and the room temperature qualification rate under different meteorological information; real-time statistically analyzing the heat load of the heat exchange station and the room temperature qualification rate under different meteorological information; and calculating the corrected comprehensive outdoor temperature using a correction algorithm. The meteorological information includes weather type, temperature, and wind speed. Based on different meteorological information and the corrected comprehensive outdoor temperature, the next control cycle searches for similar meteorological information conditions by locating the weather category and using the principle of minimizing the difference in the corrected comprehensive outdoor temperature to obtain the corresponding predicted heat load value. In step S2, the conversion correction algorithm is as follows: When the current wind speed is lower than the baseline wind speed, the calculated composite outdoor temperature T is the actual outdoor temperature t; When the current wind speed is within the baseline wind speed and the incremental wind speed threshold, the formula for calculating the converted comprehensive outdoor temperature T is as follows: ; wherein, the The baseline wind speed is... The current wind speed; When the current wind speed reaches the incremental wind speed threshold but does not exceed the medium wind speed threshold, the formula for calculating the converted comprehensive outdoor temperature T is as follows: , wherein The incremental wind speed threshold, Based on the adjustment value, , This is the relevant adjustment coefficient for the wind speed increment; When the current wind speed exceeds the medium wind speed threshold and reaches a high wind speed, the formula for calculating the equivalent comprehensive outdoor temperature T is as follows: , A fixed deduction value is preset for high wind speeds.

2. The method for continuously optimizing heating load forecasting according to claim 1, characterized in that, In step S2, the data processing module integrated on the heating platform service module performs data preprocessing on the operating data, which further includes validity processing of the operating data, including removing the operating data within a preset time after the heating network operation adjustment, and standardizing, detecting and cleaning the operating data.

3. The method for continuously optimizing heating load forecasting according to claim 1, characterized in that, In step S2, based on different meteorological information and the calculated comprehensive outdoor temperature, the next control cycle searches for similar meteorological information conditions according to the search and location weather category, and uses the principle of minimizing the difference in the calculated comprehensive outdoor temperature to obtain the corresponding predicted heat load value, further including... Set an expected value for the room temperature compliance rate to assess whether the current heating effect of the heat exchange station meets expectations; The judgment is made based on the room temperature compliance rate under the same meteorological information conditions; When the room temperature compliance rate reaches or exceeds the expected value, the predicted heat load is set as the historical heat load of the heat exchange station under similar meteorological conditions. If the room temperature compliance rate does not reach the expected value, the predicted heat load is set to the heat load of the heat exchange station under similar meteorological conditions plus an adjustment value β, where the adjustment value β is the amount of heat load used for adjustment when the room temperature compliance rate exceeds the expected value.

4. The method for continuously optimizing heating load forecasting according to claim 3, characterized in that, The similar meteorological information conditions refer to the historical heat exchange station heat load with the smallest calculated comprehensive outdoor temperature difference under the same weather type.

5. The method for continuously optimizing heating load forecasting according to claim 4, characterized in that, It also includes step S3: sending the predicted heat load value to the heating control system as the basis for adjusting the heating load; at the same time, the heating control system adjusts the operating parameters of the heat exchange station according to the received predicted heat load value and feeds back the operating data to the real-time database service module.

6. A continuous optimization heating load forecasting system based on big data analysis and learning, employing the continuous optimization heating load forecasting method as described in any one of claims 1 to 5, characterized in that, include, The real-time library service module is used to configure the data interface to receive real-time and historical operating data, including meteorological information data, heat load of heat exchange station, and room temperature qualification rate. The relationship module is used to store the analysis algorithms for analyzing the preprocessed running data; The data processing module is used to preprocess the collected operational data and analyze the preprocessed operational data using the analysis algorithm stored in the relationship module to determine the predicted heat load value for the next control cycle. Specifically, the analysis algorithm stored in the relationship module to determine the predicted heat load value for the next control cycle includes: based on the real-time collected operational data, statistically analyzing the corrected integrated outdoor temperature, the heat load of the heat exchange station, and the room temperature qualification rate under different meteorological information; based on different meteorological information and the corrected integrated outdoor temperature, the next control cycle searches for similar meteorological information conditions to obtain the corresponding predicted heat load value by searching for and locating the weather category and using the principle of minimizing the difference in the corrected integrated outdoor temperature.

7. The continuously optimized heating load prediction system according to claim 6, characterized in that, The data processing module includes, The data preprocessing unit is used to perform validity processing on the operating data, including removing the operating data within a preset time after the heating network operation adjustment, and standardizing, detecting outliers, and cleaning the operating data. The data statistics unit is used to collect the operating data in real time, and to statistically analyze the corrected comprehensive outdoor temperature, the heat load of the heat exchange station, and the room temperature qualification rate under different meteorological information. It further includes real-time statistical analysis of the heat load of the heat exchange station and the room temperature qualification rate under different meteorological information, and calculating the corrected comprehensive outdoor temperature through a conversion correction algorithm. The meteorological information includes weather type, temperature, and wind speed. The data analysis unit is used to set an expected value for the room temperature compliance rate to assess whether the current heating effect of the heat exchange station meets expectations; it makes a judgment based on the room temperature compliance rate under the same meteorological information conditions; when the room temperature compliance rate reaches or exceeds the expected value, the predicted heat load is set to the historical heat load of the heat exchange station under similar meteorological information conditions; when the room temperature compliance rate does not reach the expected value, the predicted heat load is set to the heat load of the heat exchange station under similar meteorological information conditions plus an adjustment value β, where the adjustment value β is the amount of heat load used to adjust when the room temperature compliance rate exceeds the expected value.

8. The continuously optimized heating load prediction system according to claim 7, characterized in that, It also includes a data feedback module, used to send the predicted heat load value to the heating control system as a basis for adjusting the heating load; at the same time, the heating control system adjusts the operating parameters of the heat exchange station according to the received predicted heat load value and feeds back the operating data to the real-time database service module.

Images