Equipment Adaptive Adjustment System Based on Multi-source Data Fusion

Abstract

The present invention provides an equipment adaptive adjustment system based on multi-source data fusion, which relates to the field of data processing technology. The system includes: a constraint state receiving unit, a pumping equipment updating unit, an operation data monitoring unit, a data anomaly analysis unit, a data compensation optimization unit, and an equipment adjustment execution unit. The abnormal state adaptive adjustment of the nutrient solution pumping equipment is realized by mutual calling between multiple system units, which solves the technical problem that the nutrient solution pumping equipment in the prior art can only monitor the abnormal operation of the equipment, but cannot adaptively adjust the equipment parameters according to the monitored abnormality. The technical effect of improving the control accuracy of the nutrient solution pumping equipment and avoiding the use risks caused by the abnormal control of the nutrient solution pumping equipment is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a device adaptive adjustment system based on multi-source data fusion. Background Art

[0002] At present, the nutrient solution pumping equipment is limited to monitoring equipment operation abnormalities, and is unable to realize the function of adaptively adjusting equipment parameters according to the detected abnormal conditions.

[0003] For example, assuming that an abnormal flow rate is detected in the nutrient solution pumping equipment, the current technology can only record and alarm, but cannot automatically adjust the speed or other parameters of the nutrient solution pump to deal with this abnormality.

[0004] This limitation makes it impossible for the equipment to self-adjust in abnormal situations, and human intervention is required to solve the problem, which increases labor costs and operational risks.

[0005] In summary, the nutrient solution pumping equipment in the prior art can only monitor equipment operation abnormalities, but cannot adaptively adjust equipment parameters according to the monitored abnormalities. Summary of the invention

[0006] The present application provides an equipment adaptive adjustment system based on multi-source data fusion, which is used to solve the technical problem in the prior art that nutrient solution pumping equipment can only monitor equipment operation abnormalities but cannot adaptively adjust equipment parameters according to the monitored abnormalities.

[0007] In view of the above problems, the present application provides a device adaptive adjustment system based on multi-source data fusion.

[0008] The first aspect of the present application provides an equipment adaptive adjustment system based on multi-source data fusion, the system comprising: a constraint state receiving unit, used to receive the constraint state of a nutrient solution control factor through a user terminal, wherein the constraint state of the nutrient solution control factor includes an expected flow rate interval, an expected temperature interval, an expected pumping amount interval, and an expected pumping cycle interval; a pumping equipment updating unit, used to initialize a nutrient solution pumping equipment according to the expected flow rate interval, the expected temperature interval, the expected pumping amount interval, and the expected pumping cycle interval; an operation data monitoring unit, used to obtain flow rate monitoring timing information, temperature monitoring timing information, and when the nutrient solution pumping equipment is started Pumping volume prediction information; a data anomaly analysis unit, used to perform an anomaly analysis on the flow rate monitoring timing information, the temperature monitoring timing information and the pumping volume prediction information, and generate a flow rate anomaly coefficient, a temperature anomaly coefficient and a pumping volume anomaly coefficient; a data compensation optimization unit, used to perform multi-source data compensation optimization according to the flow rate anomaly coefficient, the temperature anomaly coefficient and the pumping volume anomaly coefficient, and generate a flow rate compensation control parameter, a temperature compensation control parameter and a pumping volume compensation control parameter; an equipment adjustment execution unit, used to adaptively adjust the nutrient solution pumping equipment according to the flow rate compensation control parameter, the temperature compensation control parameter and the pumping volume compensation control parameter.

[0009] A second aspect of the present application provides a computer device, which is deployed with the system described in the first aspect.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] The system provided in the embodiment of the present application is configured to receive the constraint state of the nutrient solution control factor through a user terminal through a constraint state receiving unit, wherein the constraint state of the nutrient solution control factor includes an expected flow rate interval, an expected temperature interval, an expected pumping amount interval, and an expected pumping cycle interval; a pumping device updating unit is configured to initialize the nutrient solution pumping device according to the expected flow rate interval, the expected temperature interval, the expected pumping amount interval, and the expected pumping cycle interval; an operation data monitoring unit is configured to obtain flow rate monitoring timing information, temperature monitoring timing information, and pumping amount prediction information when the nutrient solution pumping device is started; and a data anomaly analysis unit is configured to analyze the nutrient solution pumping device and the nutrient solution pumping device. An analysis unit is used to perform an abnormal analysis on the flow rate monitoring timing information, the temperature monitoring timing information and the pumping volume prediction information, and generate a flow rate abnormality coefficient, a temperature abnormality coefficient and a pumping volume abnormality coefficient; a data compensation optimization unit is used to perform multi-source data compensation optimization based on the flow rate abnormality coefficient, the temperature abnormality coefficient and the pumping volume abnormality coefficient, and generate a flow rate compensation control parameter, a temperature compensation control parameter and a pumping volume compensation control parameter; an equipment adjustment execution unit is used to adaptively adjust the nutrient solution pumping equipment based on the flow rate compensation control parameter, the temperature compensation control parameter and the pumping volume compensation control parameter. The technical effect of improving the control accuracy of the nutrient solution pumping equipment and avoiding the use risks caused by abnormal control of the nutrient solution pumping equipment is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of the structure of the device adaptive adjustment system based on multi-source data fusion provided by this application;

[0013] Figure 2 A schematic diagram of the structure for performing multi-source data compensation optimization in the device adaptive adjustment system based on multi-source data fusion provided by the present application; DETAILED DESCRIPTION

[0014] The present application provides an equipment adaptive adjustment system based on multi-source data fusion, which is used to solve the technical problem that the nutrient solution pumping equipment in the prior art can only monitor the abnormal operation of the equipment, but cannot adaptively adjust the equipment parameters according to the monitored abnormalities. The technical effect of improving the control accuracy of the nutrient solution pumping equipment and avoiding the use risks caused by abnormal control of the nutrient solution pumping equipment is achieved.

[0015] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.

[0016] Embodiment 1

[0017] like Figure 1 As shown, the present application provides a device adaptive adjustment system based on multi-source data fusion, including:

[0018] The constraint state receiving unit A100 is used to receive the constraint state of the nutrient solution control factor through the user terminal, wherein the constraint state of the nutrient solution control factor includes the expected flow rate interval, the expected temperature interval, the expected pumping amount interval and the expected pumping cycle interval;

[0019] Specifically, the regulated device of this embodiment is a nutrient solution pumping device. The nutrient solution pumping device (an unspecified model of enteral nutrition pump) can automatically adjust the flow rate of the nutrient solution, heat the liquid, and quantitatively pump enteral nutrition solution and water into the user's intestines. At the same time, it can automatically flush the nasogastric feeding tube to ensure the hygiene and smooth operation of the system.

[0020] The constraint status receiving unit is used to receive pre-set control parameters of the nutrient solution pumping device. Specifically, the constraint status receiving unit interacts with the user end of the nutrient solution pumping device to receive the constraint status of the nutrient solution control factor, wherein the constraint status of the nutrient solution control factor includes an expected flow rate range, an expected temperature range, an expected pumping volume range and an expected pumping cycle range.

[0021] The expected flow rate interval is a preset expected range of nutrient solution flow rate, which is used to ensure that the delivery speed of the nutrient solution in the intestine is within an appropriate range; the expected temperature interval is a set expected temperature range of the nutrient solution, which is used to ensure that the temperature of the nutrient solution delivered to the intestine is suitable for absorption by the living body; the expected pumping amount interval is a preset expected range of the amount of nutrient solution pumped each time, which is used to ensure that the amount of nutrient solution delivered each time is adapted to the needs of the living body; the expected pumping cycle interval is a preset expected range of the cycle interval for pumping the nutrient solution, so as to ensure that the delivery frequency of the nutrient solution meets the needs of the user's living body.

[0022] A pumping device updating unit A200, configured to initialize a nutrient solution pumping device according to the expected flow rate interval, the expected temperature interval, the expected pumping amount interval, and the expected pumping cycle interval;

[0023] It should be understood that the pumping flow rate, temperature, pumping volume and pumping cycle of the nutrient solution should not exceed the expected flow rate range, temperature range, pumping volume range and pumping cycle range when set.

[0024] Based on this, the pumping equipment update unit is used to initialize the nutrient solution pumping equipment. Specifically, the pumping equipment update unit initializes the nutrient solution pumping equipment according to the expected flow rate range, the expected temperature range, the expected pumping volume range and the expected pumping cycle range. The purpose of equipment initialization is to ensure that the nutrient solution pumping equipment is in a suitable state before starting work to ensure the safe use of the equipment.

[0025] An operation data monitoring unit A300 is used to obtain flow rate monitoring timing information, temperature monitoring timing information and pumping amount prediction information when the nutrient solution pumping device is started;

[0026] In one embodiment, the operation data monitoring unit A300 is further used for:

[0027] A310: Construct a calculus prediction function for pumping volume:

[0028] ;

[0029] in, Characterize the pumping volume prediction information, represents the flow information at the kth moment, t is the end time of the flow rate monitoring time series information, Characterizes the flow velocity monitoring value at the kth moment, Characterizes the pipe diameter, The end time of the pumped time zone;

[0030] A320: According to the pumping amount calculus prediction function, perform a pumping amount calculus prediction based on the flow rate monitoring time series information to generate the pumping amount prediction information.

[0031] In this embodiment, the operation data monitoring unit is used to monitor the pumping flow rate and temperature of the nutrient solution pumping device and to predict the pumping volume according to the pumping flow rate.

[0032] Specifically, when the nutrient solution pumping equipment is started, the operation data monitoring unit obtains the flow rate monitoring timing information characterizing the change of the flow rate of the nutrient solution in the pumping pipeline over time, and the temperature monitoring timing information characterizing the change of the temperature of the nutrient solution in the pumping pipeline over time based on the flow rate sensor and the temperature sensor.

[0033] Furthermore, the operation data monitoring unit calculates the pumping amount prediction information representing the total pumping amount of the nutrient solution within the total time span of nutrient solution pumping according to the nutrient solution pipeline diameter and the flow rate monitoring timing information, and the specific method is as follows.

[0034] Construct a calculus prediction function for pumping volume:

[0035] ;

[0036] in, Characterize the pumping volume prediction information, Characterize the flow information at the kth moment, the flow information is the product of the cross-sectional area of ​​the pumping pipeline (calculated based on the pipeline diameter) and the flow velocity at the kth moment in the pumping pipeline, t is the end time of the flow velocity monitoring time series information, Characterizes the flow velocity monitoring value at the kth moment, Characterizes the pipe diameter, is the end time of the pumping time zone, which is the total time span for nutrient solution pumping. .

[0037] Substitute multiple instantaneous monitored flow rates at multiple moments in the flow rate monitoring time series information into the pumping amount calculus prediction function to perform pumping amount calculus prediction to generate the pumping amount prediction information.

[0038] This embodiment achieves the technical effect of obtaining pumping volume prediction information close to the actual nutrient solution pumping volume by constructing a pumping volume calculus prediction function, and indirectly achieves the technical effect of providing highly reliable reference data for subsequent pumping volume abnormality analysis.

[0039] The data anomaly analysis unit A400 is used to perform an anomaly analysis on the flow rate monitoring time series information, the temperature monitoring time series information and the pumping amount prediction information to generate a flow rate anomaly coefficient, a temperature anomaly coefficient and a pumping amount anomaly coefficient;

[0040] In one embodiment, the flow rate monitoring time series information, the temperature monitoring time series information and the pumping amount prediction information are subjected to abnormal analysis to generate a flow rate abnormality coefficient, a temperature abnormality coefficient and a pumping amount abnormality coefficient, and the data abnormality analysis unit A400 is further used to:

[0041] A410: traversing the flow rate monitoring time series information and the flow rate expected interval to perform minimum distance analysis and generate a first distance feature set;

[0042] A420: Calculate the proportion of the first distance feature set that is greater than or equal to the first distance threshold, and set it as a first flow velocity anomaly coefficient;

[0043] A430: Calculate the monitored flow velocity variance of the flow velocity monitoring time series information and set it as the second flow velocity anomaly coefficient;

[0044] A440: traversing the temperature monitoring time series information and performing minimum distance analysis on the expected temperature interval to generate a second distance feature set;

[0045] A450: Calculate the proportion of the second distance feature set that is greater than or equal to the second distance threshold, and set it as the first temperature anomaly coefficient;

[0046] A460: Calculate the monitoring temperature variance of the temperature monitoring time series information and set it as a second temperature anomaly coefficient;

[0047] A470: performing minimum distance analysis on the pumping volume prediction information and the pumping volume expected interval to generate the pumping volume abnormality coefficient;

[0048] A480: adding the first flow velocity anomaly coefficient and the second flow velocity anomaly coefficient to the flow velocity anomaly coefficient;

[0049] A490: Add the first temperature anomaly coefficient and the second temperature anomaly coefficient into the temperature anomaly coefficient.

[0050] Specifically, in this embodiment, the data anomaly analysis unit is used to quantify temperature anomalies, quantify flow rate anomalies, and quantify pumping volume anomalies. The data anomaly analysis unit specifically performs data processing for anomaly quantification as follows.

[0051] The flow rate monitoring timing information includes multiple instantaneous monitoring flow rates at multiple moments, and based on the expected flow rate interval, the multiple instantaneous monitoring flow rates are divided into a first group of instantaneous monitoring flow rates that are less than the minimum flow rate expectation of the expected flow rate interval, a second group of instantaneous monitoring flow rates that fall within the expected flow rate interval, and a third group of instantaneous monitoring flow rates that are greater than the maximum flow rate expectation of the expected flow rate interval.

[0052] The absolute values ​​of the differences between the first set of instantaneous monitored flow rates and the expected minimum flow rate, and the absolute values ​​of the differences between the third set of instantaneous monitored flow rates and the expected maximum flow rate are calculated respectively to obtain the first distance feature set including a plurality of flow rate deviation distances.

[0053] The first distance threshold is a flow rate error that allows the instantaneous monitored flow rate to deviate from the expected flow rate interval. The proportion of the amount of data in the first distance feature set that is greater than or equal to the first distance threshold in the flow rate monitoring time series information is calculated and set as the first flow rate anomaly coefficient.

[0054] The monitoring flow rate variance calculation is performed based on the multiple instantaneous monitoring flow rates of the flow rate monitoring time series information, and the variance calculation result is set as the second flow rate anomaly coefficient.

[0055] The temperature monitoring time series information includes multiple instantaneous monitoring temperatures at multiple moments, and multiple instantaneous monitoring flow rates and multiple instantaneous monitoring temperature data are aligned. The same method as that for obtaining the first distance feature set is adopted to traverse the temperature monitoring time series information and the temperature expectation interval for minimum distance analysis to generate a second distance feature set.

[0056] The second distance threshold is similar in concept to the first distance threshold. The second distance threshold is the temperature error that allows the instantaneous monitored temperature to deviate from the expected temperature range. The same method is used to calculate the first flow velocity anomaly coefficient to calculate the proportion of the second distance feature set greater than or equal to the second distance threshold, which is set as the first temperature anomaly coefficient.

[0057] The temperature variance is monitored based on a plurality of instantaneous monitored temperatures of the temperature monitoring time series information to obtain the second temperature anomaly coefficient.

[0058] The maximum and minimum values ​​of the expected pumping volume interval are called to obtain the maximum pumping volume expectation and the minimum pumping volume expectation, and it is determined whether the pumping volume prediction information is greater than the maximum pumping volume expectation or less than the minimum pumping volume expectation. If the pumping volume prediction information is greater than the maximum pumping volume expectation, the absolute value of the deviation between the pumping volume prediction information and the maximum pumping volume expectation is calculated as the pumping volume abnormality coefficient; if the pumping volume prediction information is less than the minimum pumping volume expectation, the absolute value of the deviation between the pumping volume prediction information and the minimum pumping volume expectation is calculated as the pumping volume abnormality coefficient.

[0059] The first flow velocity anomaly coefficient and the second flow velocity anomaly coefficient are added into the flow velocity anomaly coefficient; the first temperature anomaly coefficient and the second temperature anomaly coefficient are added into the temperature anomaly coefficient.

[0060] This embodiment obtains the anomaly coefficients of quantified temperature anomaly, quantified flow velocity anomaly and quantified pumping volume anomaly based on calculation, thereby achieving the technical effect of providing highly reliable data reference for subsequent equipment compensation adjustment based on the quantified anomaly.

[0061] The data compensation optimization unit A500 is used to perform multi-source data compensation optimization according to the velocity anomaly coefficient, the temperature anomaly coefficient and the pumping volume anomaly coefficient, and generate a velocity compensation control parameter, a temperature compensation control parameter and a pumping volume compensation control parameter;

[0062] In one embodiment, Figure 2 As shown, multi-source data compensation optimization is performed according to the velocity anomaly coefficient, the temperature anomaly coefficient and the pumping amount anomaly coefficient to generate velocity compensation control parameters, temperature compensation control parameters and pumping amount compensation control parameters. The data compensation optimization unit A500 is also used for:

[0063] A510: When the first flow rate abnormality coefficient is greater than or equal to the first flow rate abnormality coefficient threshold, or / and the second flow rate abnormality coefficient is greater than or equal to the second flow rate abnormality coefficient threshold, performing flow rate multi-source data compensation optimization to generate the flow rate compensation control parameter;

[0064] A520: When the first temperature anomaly coefficient is greater than or equal to a first temperature anomaly coefficient threshold, or / and the second temperature anomaly coefficient is greater than or equal to a second temperature anomaly coefficient threshold, performing temperature multi-source data compensation optimization to generate the temperature compensation control parameter;

[0065] A530: When the pumping volume abnormality coefficient is greater than or equal to the pumping volume abnormality coefficient threshold, the distance between the pumping volume prediction information and the middle value of the pumping volume expected interval is obtained, and set as the pumping volume compensation control parameter.

[0066] In one embodiment, when the first flow rate abnormality coefficient is greater than or equal to the first flow rate abnormality coefficient threshold, or / and the second flow rate abnormality coefficient is greater than or equal to the second flow rate abnormality coefficient threshold, flow rate multi-source data compensation optimization is performed to generate the flow rate compensation control parameter, and the data compensation optimization unit A500 is further used to:

[0067] A511: performing local backtracking on the nutrient solution pumping device with the nutrient solution type identifier as a constraint to obtain a first system error measurement data set;

[0068] A512: Performing online backtracking with the nutrient solution pumping equipment model and the nutrient solution type identifier as constraints to obtain a second system error measurement data set, wherein a first data volume of the second system error measurement data set is less than or equal to a second data volume of the first system error measurement data set;

[0069] A513: performing outlier removal on the first system error measurement data set and the second system error measurement data set to generate system error high-frequency measurement values, performing mean calculation on the system error high-frequency measurement values ​​to generate system error characteristic values;

[0070] A514: Generate the flow rate compensation control parameter based on the system error characteristic value.

[0071] In one embodiment, when the first temperature anomaly coefficient is greater than or equal to a first temperature anomaly coefficient threshold, or / and the second temperature anomaly coefficient is greater than or equal to a second temperature anomaly coefficient threshold, temperature multi-source data compensation optimization is performed to generate the temperature compensation control parameter, and the data compensation optimization unit A500 is further used to:

[0072] A521: Obtaining the ambient temperature monitoring value, nutrient solution type identification and heating power constraint range;

[0073] A522: using the ambient temperature monitoring value, the nutrient solution type identifier, and the expected temperature range as scene constraints, and the heating power constraint range as a variable constraint to perform local backtracking to obtain a first heating power selected value set;

[0074] A523: using the ambient temperature monitoring value, the nutrient solution type identifier, the nutrient solution pumping equipment model and the expected temperature range as scenario constraints, and the heating power constraint range as a variable constraint to perform network backtracking to obtain a second heating power selected value set;

[0075] A524: activating the temperature prediction network through the nutrient solution type identifier and the nutrient solution pumping device model, traversing the first heating power selected value set and the second heating power selected value set, and processing them in combination with the ambient temperature monitoring value to generate a first temperature prediction value set;

[0076] A525: When any one of the first temperature prediction value sets belongs to the expected temperature range, setting the associated heating power as the temperature compensation control parameter;

[0077] A526: Otherwise, after updating the first heating power selected value set and the second heating power selected value set, execute the loop.

[0078] In one embodiment, the first heating power selected value set and the second heating power selected value set are updated, and the data compensation optimization unit A500 is further used to:

[0079] A526-1: traverse the first temperature prediction value set to calculate the minimum distance to the expected temperature interval, and set it as a third distance feature set;

[0080] A526-2: sorting the first heating power selected value set and the second heating power selected value set from small to large according to the third distance feature set, and generating a heating power sorting result;

[0081] A526-3: extracting a first amount of head heating power and a second amount of tail heating power;

[0082] A526-4: Taking the first quantity of head heating power as the target, perturb the second quantity of tail heating power, and update the first heating power selected value set and the second heating power selected value set.

[0083] Specifically, in this embodiment, the data compensation optimization unit is used to analyze and determine the specific parameter adjustment compensation data of the nutrient solution pumping equipment with reference to the flow rate abnormality coefficient, the temperature abnormality coefficient and the pumping volume abnormality coefficient.

[0084] A first flow rate abnormality coefficient threshold and a second flow rate abnormality coefficient threshold are preset, and the first flow rate abnormality coefficient threshold and the second flow rate abnormality coefficient threshold are dual benchmarks for determining whether to perform current flow rate control parameter compensation.

[0085] If and only if the first flow rate abnormality coefficient is less than the first flow rate abnormality coefficient threshold and the second flow rate abnormality coefficient is less than the second flow rate abnormality coefficient threshold, it is considered that although there is a flow rate deviation at present, the deviation does not affect the pumping effect.

[0086] When the first flow rate anomaly coefficient is greater than or equal to a first flow rate anomaly coefficient threshold, or / and the second flow rate anomaly coefficient is greater than or equal to a second flow rate anomaly coefficient threshold, flow rate multi-source data compensation optimization is performed.

[0087] Specifically, before executing the flow rate multi-source data compensation optimization, the embodiment interactively obtains the nutrient solution type identifier of the nutrient solution type specifically pumped by the nutrient solution pumping device.

[0088] The nutrient solution pumping device can pump multiple types of nutrient solutions, and when each type of nutrient solution is pumped, there is a situation where the set flow rate parameters and the flow rate parameters of the actual delivery process are inconsistent, and this inconsistency is a system error.

[0089] Based on this, this embodiment performs local backtracking on the nutrient solution pumping equipment with the nutrient solution type identification as a constraint to obtain the first system error measurement data set. The first system error measurement data set is the system error data of the current nutrient solution pumping equipment as a single device in the history of pumping nutrient solution of the type corresponding to the nutrient solution type identification. The first system error measurement data set specifically includes multiple groups of set flow rate parameters-measured flow rate parameters.

[0090] Using the nutrient solution pumping equipment model and the nutrient solution type identifier as constraints, network backtracking is performed to expand the amount of retrieved data, and a second system error measurement data set is obtained. The second system error measurement data set is system error data of multiple equipment of the same type as the current nutrient solution pumping equipment for pumping nutrient solution of the type corresponding to the nutrient solution type identifier. The second system error measurement data set specifically includes multiple groups of set flow rate parameters-measured flow rate parameters for multiple pumping equipment of the same model.

[0091] It should be understood that since equipment of the same model will produce some system errors that are only applicable to its own application after being put into service, when performing online retrospective data collection, it is required that the first data volume of the second system error measurement data set obtained is less than or equal to the second data volume of the first system error measurement data set.

[0092] A rectangular coordinate system is drawn with the set value as the horizontal axis and the measured value as the vertical axis, and all the set flow rate parameters-measured flow rate parameters of the first system error measurement data set and the second system error measurement data set are substituted into the rectangular coordinate system to locate and obtain a number of scattered points.

[0093] The fitting model is selected based on the data trend of several scattered points, and then the best fitting parameters are determined using the least squares method to generate a fitting function. Then, based on the fitting function, line fitting is performed in a rectangular coordinate system, and the vertical distances between several scattered points and the fitting line are calculated to obtain several residuals.

[0094] A preset threshold is used to traverse the residuals, and points exceeding the threshold are obtained as outliers. The set flow rate parameters-measured flow rate parameters corresponding to the outliers are then deleted to generate high-frequency measurement values ​​of system errors. The high-frequency measurement values ​​of system errors are M groups of set flow rate parameters-measured flow rate parameters (M is a positive integer).

[0095] The mean of the M set flow rate parameters and the M measured flow rate parameters of the high-frequency measurement values ​​of the system error is calculated respectively to obtain the set flow rate mean and the measured flow rate mean, and then the set flow rate mean is used as the minuend and the measured flow rate mean is used as the subtrahend to obtain the difference as the system error characteristic value.

[0096] Interactively obtain the flow rate control parameters of the nutrient solution pumping equipment in the process of generating the flow rate monitoring timing information and the temperature monitoring timing information, adjust the flow rate control parameters according to the system error characteristic value, generate the flow rate compensation control parameters, and adjust the nutrient solution pumping equipment based on the flow rate compensation control parameters, so as to achieve the flow rate control of the nutrient solution pumping equipment to meet the expected flow rate range.

[0097] A first temperature anomaly coefficient threshold and a second temperature anomaly coefficient threshold are preset, and the first temperature anomaly coefficient threshold and the second temperature anomaly coefficient threshold are dual benchmarks for determining whether to perform current temperature control parameter compensation.

[0098] It should be understood that this embodiment does not impose any numerical restrictions on the various types of abnormal coefficient thresholds in the technical solution, and the numerical settings are specifically made according to the pumping accuracy requirements of the nutrient solution pumping equipment in actual applications.

[0099] If and only if the first temperature anomaly coefficient is less than the first temperature anomaly coefficient threshold, and the second temperature anomaly coefficient is less than the second temperature anomaly coefficient threshold, it is considered that although there is a temperature deviation at present, the deviation does not affect the pumping effect.

[0100] When the first temperature anomaly coefficient is greater than or equal to the first temperature anomaly coefficient threshold, or / and the second temperature anomaly coefficient is greater than or equal to the second temperature anomaly coefficient threshold, temperature multi-source data compensation optimization is performed to obtain the temperature compensation control parameters for controlling the pumping temperature of the nutrient solution pumping equipment within the expected temperature range.

[0101] The specific data processing process for obtaining the temperature compensation control parameters is as follows:

[0102] Obtain ambient temperature monitoring values, nutrient solution type identification and heating power constraint intervals. The ambient temperature monitoring values ​​are multiple instantaneous ambient temperature data recorded by the environmental monitoring sensor at multiple moments in the process of obtaining flow rate monitoring timing information, temperature monitoring timing information and pumping volume prediction information. The heating power constraint interval is the heating power control constraint of the nutrient solution pumping equipment for preheating the nutrient solution. It should be understood that the heating power affects the nutrient solution temperature.

[0103] The local database of the nutrient solution pumping device stores the historical heating powers for heating various nutrient solutions to different temperatures under various ambient temperatures, that is, multiple groups of historical ambient temperature monitoring values-historical nutrient solution types-historical nutrient solution temperatures-historical heating powers.

[0104] The maximum and minimum values ​​of the ambient temperature monitoring value are called to obtain the ambient temperature constraint interval, the ambient temperature constraint interval, the nutrient solution type identifier and the expected temperature interval are used as scene constraints, and the heating power constraint interval is used as a variable constraint to perform local backtracking to call data on the local database to obtain a first heating power selected value set, the first heating power selected value set including M groups of historical ambient temperature means (obtained by calculating the mean of historical ambient temperature monitoring values) - historical nutrient solution types - historical nutrient solution temperatures - historical heating powers, and each group of historical ambient temperature means - historical nutrient solution temperatures - historical heating powers in the first heating power selected value set meets the ambient temperature constraint interval, the nutrient solution type identifier, the expected temperature interval and the heating power constraint interval.

[0105] The same method is adopted, with the ambient temperature constraint interval, the nutrient solution type identifier, the nutrient solution pumping equipment model and the expected temperature interval as scene constraints, and the heating power constraint interval as a variable constraint for network backtracking, so as to obtain H groups of historical ambient temperature mean values ​​- historical nutrient solution temperature - historical heating power that meet the ambient temperature constraint interval, the nutrient solution type identifier, the expected temperature interval and the heating power constraint interval in the local database of the same model of equipment as the nutrient solution pumping equipment, to constitute the second heating power selected value set.

[0106] The temperature prediction network is activated through the nutrient solution type identifier and the nutrient solution pumping equipment model, so that the temperature prediction network can be directed to perform nutrient solution pumping temperature prediction based on ambient temperature and heating power when a specific model of nutrient solution pumping equipment is pumping a specific type of nutrient solution.

[0107] The temperature prediction network is a back-propagation neural network, which uses the H+M groups of historical ambient temperature average-historical nutrient solution temperature-historical heating power of the first heating power selected value set and the second heating power selected value set as training data, and uses the conventional training method of the back-propagation neural network to train the temperature prediction network until the nutrient solution temperature prediction accuracy of the temperature prediction network is stably higher than 98.6%.

[0108] The first heating power selected value set and the second heating power selected value set are traversed to perform all calls of historical heating powers to obtain a first heating power value set, wherein the first heating power value set includes H+M historical heating powers.

[0109] The first heating power value set is combined with the environmental temperature monitoring value and input into the temperature prediction network to predict the nutrient solution heating temperature, thereby generating a first temperature prediction value set, wherein the first temperature prediction value set includes H+M temperature prediction values.

[0110] When any one of the first temperature prediction value sets belongs to the expected temperature interval, the associated heating power obtained by mapping and calling the first heating power value set is set as the temperature compensation control parameter.

[0111] Otherwise, the same method as that for obtaining the first distance feature set is adopted to traverse the first temperature prediction value set to calculate the minimum distance to the expected temperature interval, and set it as the third distance feature set.

[0112] According to the mapping relationship between the H+M temperature distance features of the third distance feature set and the H+M historical heating powers, and taking the historical heating power as a sorting reference, the first heating power selected value set and the second heating power selected value set are sorted from small to large to generate a heating power sorting result, which includes a first heating power sequence and a second heating power sequence.

[0113] A first quantity of head heating power is extracted from the first heating power sequence, and a second quantity of tail heating power is extracted from the second heating power sequence, wherein the heating power data amounts of the first quantity of head heating power and the second quantity of tail heating power are consistent.

[0114] Perform mean calculation on the first quantity of head heating power and the second quantity of tail heating power to obtain the second quantity of disturbance heating power, and use the second quantity of disturbance heating power to replace the second quantity of tail heating power in the second heating power selected value set, and add the second quantity of tail heating power to the first heating power selected value set, thereby completing the update of the first heating power selected value set and the second heating power selected value set.

[0115] Afterwards, the temperature prediction network is used to traverse the updated first heating power selected value set and the second heating power selected value set in combination with the ambient temperature monitoring value to perform a cycle of nutrient solution temperature prediction and temperature expected range comparison until the associated heating power falling within the expected temperature range is obtained as the temperature compensation control parameter.

[0116] When the pumping volume abnormality coefficient is greater than or equal to the pumping volume abnormality coefficient threshold, the distance between the pumping volume prediction information and the middle value of the expected pumping volume interval is obtained and set as the pumping volume compensation control parameter. Otherwise, there is no need to perform pumping volume control compensation.

[0117] This embodiment achieves the technical effect of analyzing and determining the nutrient solution pumping temperature, pumping flow rate and pumping amount of the nutrient solution pumping equipment to eliminate system errors and meet the expected flow rate range, temperature range and pumping amount range.

[0118] The equipment adjustment execution unit A600 is used to adaptively adjust the nutrient solution pumping equipment according to the flow rate compensation control parameter, the temperature compensation control parameter, and the pumping amount compensation control parameter.

[0119] In one embodiment, the equipment adjustment execution unit is an actual execution unit of the compensation control, which adaptively adjusts the nutrient solution pumping equipment according to the flow rate compensation control parameters, the temperature compensation control parameters, and the pumping amount compensation control parameters, so that the nutrient solution pumping temperature, pumping flow rate, and pumping amount of the nutrient solution pumping equipment can eliminate system errors and meet the expected flow rate range, temperature range, and pumping amount range.

[0120] This embodiment achieves the technical effect of improving the control accuracy of the nutrient solution pumping device and avoiding the use risks caused by abnormal control of the nutrient solution pumping device.

[0121] Embodiment 2

[0122] Based on the same inventive concept as the device adaptive adjustment system based on multi-source data fusion in the aforementioned embodiment, the present application also provides a computer device, which is deployed with the system described in the first embodiment.

[0123] Based on the above specific embodiments of the present invention, any improvements and modifications made to the present invention by those skilled in the art without departing from the principles of the present invention shall fall within the patent protection scope of the present invention.

Claims

1. Equipment adaptive adjustment system based on multi-source data fusion, It is characterized in that include: A constraint state receiving unit, used to receive the constraint state of the nutrient solution control factor through the user terminal, wherein the constraint state of the nutrient solution control factor includes an expected flow rate interval, an expected temperature interval, an expected pumping amount interval, and an expected pumping cycle interval; A pumping device updating unit, configured to initialize a nutrient solution pumping device according to the expected flow rate interval, the expected temperature interval, the expected pumping amount interval, and the expected pumping cycle interval; An operation data monitoring unit, used to obtain flow rate monitoring timing information, temperature monitoring timing information and pumping amount prediction information when the nutrient solution pumping device is started; A data anomaly analysis unit, used for performing an anomaly analysis on the flow rate monitoring time series information, the temperature monitoring time series information and the pumping amount prediction information, and generating a flow rate anomaly coefficient, a temperature anomaly coefficient and a pumping amount anomaly coefficient; A data compensation optimization unit, used for performing multi-source data compensation optimization according to the velocity anomaly coefficient, the temperature anomaly coefficient and the pumping volume anomaly coefficient, to generate a velocity compensation control parameter, a temperature compensation control parameter and a pumping volume compensation control parameter; An equipment adjustment execution unit, used for adaptively adjusting the nutrient solution pumping equipment according to the flow rate compensation control parameter, the temperature compensation control parameter, and the pumping amount compensation control parameter; The data anomaly analysis unit is used for: Traversing the flow rate monitoring time series information and the flow rate expected interval to perform minimum distance analysis to generate a first distance feature set; Calculate the proportion of the first distance feature set that is greater than or equal to the first distance threshold, and set it as a first flow velocity anomaly coefficient; Calculate the monitored flow velocity variance of the flow velocity monitoring time series information and set it as a second flow velocity anomaly coefficient; Traversing the temperature monitoring time series information and the temperature expected interval to perform minimum distance analysis to generate a second distance feature set; Calculate the proportion of the second distance feature set that is greater than or equal to the second distance threshold, and set it as the first temperature anomaly coefficient; Calculate the monitoring temperature variance of the temperature monitoring time series information and set it as a second temperature anomaly coefficient; Performing minimum distance analysis on the pumping volume prediction information and the pumping volume expected interval to generate the pumping volume abnormality coefficient; adding the first flow velocity anomaly coefficient and the second flow velocity anomaly coefficient into the flow velocity anomaly coefficient; adding the first temperature anomaly coefficient and the second temperature anomaly coefficient into the temperature anomaly coefficient; The data compensation optimization unit is used for: When the first flow rate abnormality coefficient is greater than or equal to a first flow rate abnormality coefficient threshold, or / and the second flow rate abnormality coefficient is greater than or equal to a second flow rate abnormality coefficient threshold, performing flow rate multi-source data compensation optimization to generate the flow rate compensation control parameter; When the first temperature anomaly coefficient is greater than or equal to a first temperature anomaly coefficient threshold, or / and the second temperature anomaly coefficient is greater than or equal to a second temperature anomaly coefficient threshold, performing temperature multi-source data compensation optimization to generate the temperature compensation control parameter; When the pumping amount abnormality coefficient is greater than or equal to the pumping amount abnormality coefficient threshold, the distance between the pumping amount prediction information and the middle value of the pumping amount expected interval is obtained and set as the pumping amount compensation control parameter; The data compensation optimization unit is used for: Performing local backtracking on the nutrient solution pumping device with the nutrient solution type identifier as a constraint to obtain a first system error measurement data set; Performing online backtracking with the nutrient solution pumping equipment model and the nutrient solution type identifier as constraints to obtain a second system error measurement data set, wherein a first data volume of the second system error measurement data set is less than or equal to a second data volume of the first system error measurement data set; Deleting outliers from the first system error measurement data set and the second system error measurement data set to generate system error high-frequency measurement values, and performing mean calculation on the system error high-frequency measurement values ​​to generate system error characteristic values; generating the flow rate compensation control parameter according to the system error characteristic value; The data compensation optimization unit is used for: Obtaining the ambient temperature monitoring value, nutrient solution type identification and heating power constraint range; Taking the ambient temperature monitoring value, the nutrient solution type identifier and the expected temperature range as scene constraints and the heating power constraint range as a variable constraint, local backtracking is performed to obtain a first heating power selected value set; Taking the ambient temperature monitoring value, the nutrient solution type identifier, the nutrient solution pumping equipment model and the expected temperature range as scenario constraints, and taking the heating power constraint range as a variable constraint, network backtracking is performed to obtain a second heating power selected value set; Activate the temperature prediction network through the nutrient solution type identifier and the nutrient solution pumping equipment model, traverse the first heating power selected value set and the second heating power selected value set, and process them in combination with the ambient temperature monitoring value to generate a first temperature prediction value set; When any one of the first temperature prediction value sets belongs to the expected temperature range, setting the associated heating power as the temperature compensation control parameter; Otherwise, after updating the first heating power selected value set and the second heating power selected value set, a loop is executed.

2. The system according to claim 1, It is characterized in that The operation data monitoring unit is used for: Construct a calculus prediction function for pumping volume: ； in, Characterize the pumping volume prediction information, represents the flow information at the kth moment, t is the end time of the flow rate monitoring time series information, Characterizes the flow velocity monitoring value at the kth moment, Characterizes the pipe diameter, The end time of the pumped time zone; According to the pumping amount calculus prediction function, a pumping amount calculus prediction is performed based on the flow rate monitoring time series information to generate the pumping amount prediction information.

3. The system according to claim 1, It is characterized in that The data compensation optimization unit is used for: Traversing the first temperature prediction value set to calculate the minimum distance to the expected temperature interval, and setting it as a third distance feature set; According to the third distance feature set, the first heating power selected value set and the second heating power selected value set are sorted from small to large to generate a heating power sorting result; Extracting a first amount of head heating power and a second amount of tail heating power; Taking the first number of head heating powers as a target, the second number of tail heating powers is disturbed, and the first heating power selected value set and the second heating power selected value set are updated.

4. A computer device, It is characterized in that A system as claimed in any one of claims 1 to 3 is deployed.

Images