Distributed energy station heat supply load prediction method based on LSTM

Abstract

The invention discloses a distributed energy station heat supply load prediction method based on LSTM, relates to the technical field of distributed energy station heat supply, and provides a distributed energy station heat supply load prediction method based on a long short-term memory network. Through learning past historical operation data, weather data, building data and time data of each distributed energy station after abnormal processing of an isolated forest algorithm, a unique memory mechanism and a gating structure of LSTM are applied to analyze a heat supply load change rule of the energy station. And the most reasonable outlet water temperature or return water temperature in a certain hour in the future is predicted according to real-time data and is issued to the energy station in time, the heating temperature of the air source heat pump is adjusted, accurate heat supply regulation and control are achieved, room temperature balance can be effectively maintained, the COP value of the energy station is increased, and work orders and energy waste are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed energy station heating, and specifically to a distributed energy station heating load prediction method based on LSTM. Background Art

[0002] Distributed energy station heating refers to a heating method in which heat source equipment is placed near the user and heat energy is directly delivered to the user's equipment through the community heating network or the internal pipes of the building. In order to maintain a constant heating user experience and to avoid energy waste to a certain extent, the distributed energy station should adjust the heating load value in a timely manner according to factors such as weather changes, equipment conditions, and changes in heating area. At present, the load adjustment method of distributed energy stations mainly relies on manual experience or configuration settings based on manual experience. The accuracy and efficiency of the adjustment process largely depends on the operator's professional knowledge, accumulated experience, and keen insight into the actual situation.

[0003] Therefore, there are certain limitations in load regulation of distributed energy station heating: the stations are scattered, and the conditions of different stations vary greatly, making it difficult to conduct unified regulation and control. If decisions are made based on manual experience, it is difficult to allocate enough manpower to achieve accurate regulation of each station, and the labor cost is high; the adjustment frequency is low, the adjustment is not timely, and it is impossible to respond to subtle changes in actual load in time, resulting in insufficient or excessive heating from time to time; the adjustment is overly dependent on manual expertise. Summary of the invention

[0004] The purpose of the present invention is to provide a distributed energy station heating load prediction method based on LSTM to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for predicting heating load of a distributed energy station based on LSTM, the method comprising the following steps: S1. Obtain historical data of a distributed energy station, perform data missing processing based on the historical data, and record the number of missing processing; S2. Build an isolation forest model based on historical data to determine abnormal data, perform data missing processing on the abnormal data, and record it in the missing processing quantity; S3. For the processed data, a distributed energy station load forecasting model is constructed based on the long short-term memory network; S4. Build a system evaluation model to update and reuse the distributed energy station load prediction model based on the real-time operation data of the distributed energy station load.

[0006] According to the above technical solution, the historical data includes: site operation data, weather data, building data and time data; The site operation data includes supply water temperature, return water temperature, instantaneous heat, instantaneous flow, accumulated heat, accumulated electricity, indoor temperature and target room temperature; the weather data includes temperature, humidity, rainfall, wind speed, wind direction, weather type and solar radiation; the building data includes floor, heating area, building insulation level, building type and heating type; the time data includes the hour of the day, the week and whether it is a holiday.

[0007] According to the above technical solution, the data missing processing includes: When historical data is missing, choose to abandon or linear interpolate the processing method, including: If there are complete data records before and after the missing data point, the missing data is supplemented by linear interpolation. Linear interpolation estimates the value of the missing point by drawing a straight line based on the known data before and after the missing data point. The drawing method comprises: Data exists before the missing point , data exists after the missing point , the time difference between the two existing data is , the time difference between the missing point and the data before the missing point is , then the missing point data is:

[0008] If the data before and after the missing data point are incomplete, or if multiple data points are missing consecutively, choose to discard the missing data point.

[0009] According to the above technical solution, the abnormal data determination includes: The data is grouped and similar data is grouped together. The grouping process specifically includes: Weather data: Weather codes and outdoor temperature are used as the grouping basis. Weather codes are used to reflect changes in rainfall, humidity, and solar radiation. Outdoor temperature is divided into groups of 5 degrees Celsius, and the specific endpoints include the front but not the back. Building data: take the number of floors as the basis for grouping; Time data: divided by the time range of the day, whether it is a weekend, whether it is a holiday, where the time range of the day is divided into daytime working hours and nighttime off-get off work time; the "daytime working hours" and "nighttime off-get off work time" are set according to the specific conditions of different sites, and the default refers to 7 am to 7 pm and 7 pm to 7 am, where the endpoints are included before but not after.

[0010] According to the above technical solution, it also includes: S5-1. For any group, randomly select N groups of heating operation data as subsamples and put them into the root node of an isolated tree; S5-2. At each root node, randomly select a feature in the heating operation data, and randomly select a value between the maximum and minimum values ​​of the feature as the split value. The data is divided into the left subtree or the right subtree according to the split value; S5-3, randomly select one of the other data items of the heating operation data as a feature, and repeat step S5-2 until the data cannot be divided any further, and an isolated tree is obtained; S5-4, repeat steps S5-1 to S5-3 to obtain M isolated trees to form an isolation forest, where M is the number of isolated trees obtained by applying the isolation forest algorithm to the data in one group; S5-5. In the isolation forest, a threshold value of the number of split points is set to judge any data. During the isolation process, if the number of split points exceeds the threshold value, it is judged as normal data; if the number of split points does not exceed the threshold value, it is judged as abnormal data. S5-6. Feedback the detected abnormal data for data missing processing.

[0011] According to the above technical solution, the distributed energy station load prediction model includes: Three gating units, namely input gate, forget gate and output gate, are introduced to control the dimension of retaining the heating data at the previous moment. The LSTM input data includes: water supply temperature, return water temperature, instantaneous heat, instantaneous flow, indoor temperature, target room temperature, temperature, humidity, rainfall, wind speed, wind direction, weather type, solar radiation, floor, hour of the day, week of the week and whether it is a holiday. The forget gate determines the data information that LSTM discards from the heating data of the previous moment, specifically including: Set the current time to t, and the forget gate reads the output of the previous layer and current input real-time data , calculate the retention factor of the information at the previous moment :

[0012] Where: is the forget gate weight matrix, is the forget gate bias, is the sigmoid function, Mapped to the interval [0,1]; where Generated by the input at the previous moment and the retained information at the previous moment; The cell state obtained from the information of the previous moment and Multiply, A value of 1 indicates that All information in the , A value of 0 means completely discarding Information in The input gate is used to screen information and determine the new heating data input at the current moment to be used to update the final output data information, including a sigmoid layer and a tanh layer; The sigmoid layer is based on the output of the previous layer. and current input real-time data Calculate the update factor of the heating information at the new time :

[0013] in: is the input gate sigmoid layer weight matrix, Bias for the input gate sigmoid layer; The tanh layer calculates candidate cell states , is added to the cell state at the current moment middle:

[0014] in: is the input gate tanh layer weight matrix, is the input gate tanh layer bias, is the tanh activation function; The current cell state It is equal to the sum of the product of the output value of the forget gate and the product of the two output values ​​of the input gate, that is:

[0015] The output gate is used to calculate the output of the current layer ; The output gate first passes through a sigmoid layer to determine the cell state at the current moment. Output data information:

[0016] Where: is the output gate weight matrix, Bias for the output gate; The current cell state After processing by tanh, a value in the interval [-1,1] is obtained and it is combined with Multiply and finally determine the output :

[0017] If the current layer is the last layer, then It is the final prediction value of the distributed energy station heating load prediction model.

[0018] According to the above technical solution, it also includes: Set the prediction time span p and reference data time span k for the LSTM algorithm; Assume that the current time is d, and use k data from time d-k+1 to time d to predict the energy station load value at the future time p; When LSTM is required to provide the recommended load value at the current moment: in the model training stage, set p = 0, the training set is a group of k data, the target value of each group of data is the load value corresponding to the kth data, the weather data is the weather data at the corresponding moment of each data, and in the model inference stage, select the historical k-1 data and the current real-time data, combine the weather data at the corresponding moment, and apply the model to obtain the recommended load value at the current moment; When LSTM is needed to predict the recommended load value at a future time, set the future time R. In the model training phase, set p = Rd. The training set consists of k data items. The target value of each group of data is the load value corresponding to the k+pth data item. The weather data is the weather data at the time + p corresponding to each data item. In the model inference phase, select the historical k-1 data items and the current real-time data, combine them with the weather data at the time + p corresponding to each data item, and apply the model to predict the recommended load value at the future time R.

[0019] According to the above technical solution, the system evaluation model includes: Obtain the formed distributed energy station heating load forecasting model and the amount of missing data in the process of building the model; Define the model prediction tolerance threshold, and perform operation analysis on the formed distributed energy station heating load prediction model, wherein the operation analysis refers to: Starting from the current moment, obtain the absolute value of the difference between the actual result and the predicted result each time, and continue to accumulate until the total absolute value of the difference exceeds the model prediction tolerance threshold for the first time, and record the total number of predicted data experienced from the current moment; Taking the number of missing data as the horizontal coordinate and the total number of experienced forecast data as the vertical coordinate, linear fitting is performed on the coordinate points formed by the heating load forecasting model of all distributed energy stations to form a functional relationship between the number of missing data and the total number of experienced forecast data; Obtain the number of missing data in the current distributed energy station heating load forecasting model, feedback the total number of forecast data experienced based on the functional relationship, mark the position, and when the model reaches the marked position, feedback to the management port to remind the model to update.

[0020] According to the above technical solution, it also includes: When a new site is put into use, it is compared with the building attribute data of the site with an existing heating load prediction model, and the sites that belong to the same group with its building attributes are calculated and the model is reused. After the new site is put into use for some time, the model is updated according to the historical operation data of the new site so that the mean room temperature of the site reaches a stable state.

[0021] Compared with the prior art, the beneficial effects of the present invention are: the present invention can customize and generate its unique load forecasting model according to the historical operation data and building data of different sites, greatly improving the accuracy of regulation. When the weather changes, the system can respond quickly and adjust the load to ensure that the average room temperature remains stable, reduce the number of work orders and effectively prevent energy waste. During special periods, such as nights, weekends or holidays, the system can implement specific temperature control strategies to further reduce energy consumption. At the same time, it can also intelligently manage the start-up time of the heat pump to avoid energy waste caused by premature start-up and heating. The prediction time span can be flexibly adjusted, and the load value can be predicted a few hours later, making the regulation time more flexible. Reduce dependence on manual operation, reduce labor costs and improve overall operation and maintenance efficiency. When a new site is put into operation, there is no need to spend a lot of time on manual adjustment and optimization. The model can reuse data from other similar sites and combine the current site operation data to continuously update the model so that the site room temperature average reaches a stable state faster. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of a method for predicting heating load of a distributed energy station based on LSTM in the present invention; Figure 2 It is a schematic diagram of the LSTM structure of a distributed energy station heating load prediction method based on LSTM in the present invention; Figure 3 This is a schematic diagram of a forget gate in an embodiment of a method for predicting heating load of a distributed energy station based on LSTM according to the present invention; Figure 4 A schematic diagram of an input gate of an embodiment of a distributed energy station heating load prediction method based on LSTM of the present invention; Figure 5 It is a schematic diagram of an output gate of a distributed energy station heating load prediction method based on LSTM in the present invention; Figure 6 This is a schematic diagram of the results of an embodiment of a method for predicting heating load of a distributed energy station based on LSTM in the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0024] Example: Figure 1-Figure 6 As shown, the present invention provides a distributed energy station heating load prediction method based on LSTM: the method comprises the following steps: S1. Obtain historical data of a distributed energy station, perform data missing processing based on the historical data, and record the number of missing processing; S2. Build an isolation forest model based on historical data to determine abnormal data, perform data missing processing on the abnormal data, and record it in the missing processing quantity; S3. For the processed data, a distributed energy station load forecasting model is constructed based on the long short-term memory network; S4. Build a system evaluation model to update and reuse the distributed energy station load prediction model based on the real-time operation data of the distributed energy station load.

[0025] The historical data includes: site operation data, weather data, building data and time data; The site operation data includes supply water temperature, return water temperature, instantaneous heat, instantaneous flow, accumulated heat, accumulated electricity, indoor temperature and target room temperature; the weather data includes temperature, humidity, rainfall, wind speed, wind direction, weather type and solar radiation; the building data includes floor, heating area, building insulation level, building type and heating type; the time data includes the hour of the day, the week and whether it is a holiday.

[0026] The data missing processing includes: When historical data is missing, choose to abandon or linear interpolate the processing method, including: If there are complete data records before and after the missing data point, the missing data is supplemented by linear interpolation. Linear interpolation estimates the value of the missing point by drawing a straight line based on the known data before and after the missing data point. The drawing method comprises: For example, in n-dimensional data, there is data before the missing point. , data exists after the missing point , the time difference between the two existing data is , the time difference between the missing point and the data before the missing point is , then the missing point data is:

[0027] If the data before and after the missing data point are incomplete, or if multiple data points are missing consecutively, choose to discard the missing data point to ensure the authenticity and reliability of the data.

[0028] For some data that is difficult to judge manually whether it is abnormal, the isolation forest algorithm is used to screen out abnormalities. Isolation forest is an unsupervised anomaly detection method based on random binary trees. The heating abnormal data detection process based on isolation forest is as follows: Since different weather, building, and time data will lead to large differences in heating operation data, it is necessary to group and judge anomalies. All heating operation data are divided according to the range of weather, building, and time data, and similar data are grouped together. Each group is detected using the isolation forest algorithm to detect whether there are abnormal data in the operation data.

[0029] The grouping process specifically includes: Weather data: Weather codes and outdoor temperature are used as the grouping basis. Weather codes are used to reflect changes in rainfall, humidity, and solar radiation. Outdoor temperature is divided into groups of 5 degrees Celsius, with the specific endpoints including the front and excluding the back. For example, the outdoor temperature in [0,5) degrees Celsius is grouped as one group, and [5,10) degrees Celsius is grouped as another group. Building data: Since the heating area, building insulation level, building type, and heating type of the same site have the same values, the number of floors is used as the basis for grouping; Time data: divided by the time period of the day, whether it is a weekend, or whether it is a holiday. The time period of the day is divided into "daytime working hours" and "nighttime off-duty time". The "daytime working hours" and "nighttime off-duty time" are set according to the specific conditions of different sites, and the default refers to 7 am to 7 pm and 7 pm to 7 am, where the endpoints are included before but not after.

[0030] S5-1. For any group, randomly select N groups of heating operation data as subsamples and put them into the root node of an isolated tree; S5-2. At each root node, randomly select a feature (such as water supply temperature) in the heating operation data, and randomly select a value between the maximum and minimum values ​​of the feature as the split value. The data is divided into the left subtree or the right subtree according to this split value; S5-3, randomly select one of the other data items of the heating operation data as a feature, and repeat step S5-2 until the data cannot be divided any further, and an isolated tree is obtained; S5-4, repeat steps S5-1 to S5-3 to obtain M isolated trees to form an isolation forest, where M is the number of isolated trees obtained by applying the isolation forest algorithm to the data in one group; S5-5. In the isolation forest, a threshold value of the number of split points is set to judge any data. During the isolation process, if the number of split points exceeds the threshold value, it is judged as normal data; if the number of split points does not exceed the threshold value, it is judged as abnormal data. S5-6. Feedback the detected abnormal data for data missing processing.

[0031] Due to the time delay of heating data, the predicted load value will not affect the room temperature immediately after it is issued. Instead, it will take some time for the heat to be transferred to each room and affect the room temperature. However, due to the differences in the heating area, number of floors, etc. of each distributed energy station, the speed of heat transfer is not the same, and time cannot be uniformly aligned. Therefore, when making load forecasts, it is necessary not only to refer to the real-time heating data at the current moment, but also to comprehensively consider the heating data over the past period of time.

[0032] The long short-term memory network is used to realize the load forecasting of distributed energy stations. By introducing three gate control units, namely input gate, forget gate and output gate, we can control which dimensions of the heating data at the previous moment are retained and which dimensions of the latest real-time data are selected to update the information to obtain the final forecast result. The unit structure of the model at a certain moment is as follows: Figure 2 shown.

[0033] The distributed energy station load prediction model includes: Three gating units, namely input gate, forget gate and output gate, are introduced to control the dimension of retaining the heating data at the previous moment. The LSTM input data includes: water supply temperature, return water temperature, instantaneous heat, instantaneous flow, indoor temperature, target room temperature, temperature, humidity, rainfall, wind speed, wind direction, weather type, solar radiation, floor, hour of the day, week of the week and whether it is a holiday. like Figure 3 As shown, the forget gate determines the data information that LSTM discards from the heating data at the previous moment, specifically including: Set the current time to t, and the forget gate reads the output of the previous layer and current input real-time data , calculate the retention factor of the information at the previous moment :

[0034] Where: is the forget gate weight matrix, is the forget gate bias, is the sigmoid function, Mapped to the interval [0,1]; where Generated by the input at the previous moment and the retained information at the previous moment; The cell state obtained from the information of the previous moment and Multiply, A value of 1 indicates that All information in the , A value of 0 means completely discarding Information in like Figure 4 As shown, the input gate is used to screen information and determine the new heating data input at the current moment to be used to update the final output data information, including a sigmoid layer and a tanh layer; The sigmoid layer is based on the output of the previous layer. and current input real-time data Calculate the update factor of the heating information at the new time :

[0035] in: is the input gate sigmoid layer weight matrix, Bias for the input gate sigmoid layer; The tanh layer calculates candidate cell states , is added to the cell state at the current moment middle:

[0036] in: is the input gate tanh layer weight matrix, is the input gate tanh layer bias, is the tanh activation function; The current cell state It is equal to the sum of the product of the output value of the forget gate and the product of the two output values ​​of the input gate, that is:

[0037] like Figure 5 As shown, the output gate is used to calculate the output of the current layer ; The output gate first passes through a sigmoid layer to determine the cell state at the current moment. Output data information:

[0038] Where: is the output gate weight matrix, Bias for the output gate; The current cell state After processing by tanh, a value in the interval [-1,1] is obtained and it is combined with Multiply and finally determine the output :

[0039] If the current layer is the last layer, then It is the final prediction value of the distributed energy station heating load prediction model.

[0040] Also includes: Set the prediction time span p and reference data time span k for the LSTM algorithm; Assume that the current time is d, and use k data from time d-k+1 to time d to predict the energy station load value at the future time p; When LSTM is required to provide the recommended load value at the current moment: in the model training stage, set p = 0, the training set is a group of k data, the target value of each group of data is the load value corresponding to the kth data, the weather data is the weather data at the corresponding moment of each data, and in the model inference stage, select the historical k-1 data and the current real-time data, combine the weather data at the corresponding moment, and apply the model to obtain the recommended load value at the current moment; When LSTM is needed to predict the recommended load value at a future time, set the future time R. In the model training phase, set p = Rd. The training set consists of k data items. The target value of each group of data is the load value corresponding to the k+pth data item. The weather data is the weather data at the time + p corresponding to each data item. In the model inference phase, select the historical k-1 data items and the current real-time data, combine them with the weather data at the time + p corresponding to each data item, and apply the model to predict the recommended load value at the future time R.

[0041] In this embodiment, a heating load prediction model of a distributed energy station is selected, and the prediction results of 50 groups of data are compared with the actual data. The results are as follows: Figure 6 , where the gray solid line is the real data, the black dotted line is the prediction result, the x-axis is 50 sets of data, and the y-axis is the load prediction value; Select MSE and MAE as evaluation indicators, and the calculation formulas of MSE and MAE are:

[0042]

[0043] In the formula, is the load forecast value of the i-th group of data, is the true load value of the i-th group of data.

[0044] Before starting training, the training set and test set were divided into training set and test set in a ratio of 9:1. The MSE calculation result of the model test set was 0.08 and MAE was 0.2. The error with historical experience was small and the model prediction effect was ideal.

[0045] The system evaluation model includes: After the model has been running for a period of time, it may encounter new weather data that is significantly different from the historical data used in previous training, or user needs may change. In order to improve the generalization of the model, the model needs to be updated, that is, all previous steps are re-executed to generate a new model. The system evaluation model includes: Obtain the formed distributed energy station heating load forecasting model and the amount of missing data in the process of building the model; Define the model prediction tolerance threshold, and perform operation analysis on the formed distributed energy station heating load prediction model, wherein the operation analysis refers to: Starting from the current moment, obtain the absolute value of the difference between the actual result and the predicted result each time, and continue to accumulate until the total absolute value of the difference exceeds the model prediction tolerance threshold for the first time, and record the total number of predicted data experienced from the current moment; Taking the number of missing data as the horizontal coordinate and the total number of experienced forecast data as the vertical coordinate, linear fitting is performed on the coordinate points formed by the heating load forecasting model of all distributed energy stations to form a functional relationship between the number of missing data and the total number of experienced forecast data; Obtain the number of missing data in the current distributed energy station heating load forecasting model, feedback the total number of forecast data experienced based on the functional relationship, mark the position, and when the model reaches the marked position, feedback to the management port to remind the model to update.

[0046] When a new site is put into use, it is compared with the building attribute data of the site with an existing heating load prediction model, and the sites that belong to the same group with its building attributes are calculated and the model is reused. After the new site is put into use for some time, the model is updated according to the historical operation data of the new site so that the mean room temperature of the site reaches a stable state.

[0047] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A distributed energy station heating load prediction method based on LSTM, characterized by: The method comprises the following steps: S1. Obtain historical data of a distributed energy station, perform data missing processing based on the historical data, and record the number of missing processing; S2. Build an isolation forest model based on historical data to determine abnormal data, perform data missing processing on the abnormal data, and record it in the missing processing quantity; S3. For the processed data, a distributed energy station load forecasting model is constructed based on the long short-term memory network; S4. Build a system evaluation model to update and reuse the distributed energy station load prediction model based on the real-time operation data of the distributed energy station load.

2. According to a LSTM-based distributed energy station heating load prediction method according to claim 1, it is characterized by: The historical data includes: site operation data, weather data, building data and time data; The site operation data includes supply water temperature, return water temperature, instantaneous heat, instantaneous flow, accumulated heat, accumulated electricity, indoor temperature and target room temperature; the weather data includes temperature, humidity, rainfall, wind speed, wind direction, weather type and solar radiation; the building data includes floor, heating area, building insulation level, building type and heating type; the time data includes the hour of the day, the week and whether it is a holiday.

3. The method for predicting heating load of a distributed energy station based on LSTM according to claim 1 is characterized in that: The data missing processing includes: When historical data is missing, choose to abandon or linear interpolate the processing method, including: If there are complete data records before and after the missing data point, the missing data is supplemented by linear interpolation. Linear interpolation estimates the value of the missing point by drawing a straight line based on the known data before and after the missing data point. The drawing method comprises: Data exists before the missing point , data exists after the missing point , the time difference between the two existing data is , the time difference between the missing point and the data before the missing point is , then the missing point data is: If the data before and after the missing data point are incomplete, or if multiple data points are missing consecutively, choose to discard the missing data point.

4. The method for predicting heating load of a distributed energy station based on LSTM according to claim 3 is characterized in that: The abnormal data determination includes: The data is grouped and similar data is grouped together. The grouping process specifically includes: Weather data: Weather codes and outdoor temperature are used as the grouping basis. Weather codes are used to reflect changes in rainfall, humidity, and solar radiation. Outdoor temperature is divided into groups of 5 degrees Celsius, and the specific endpoints include the front but not the back. Building data: take the number of floors as the basis for grouping; Time data: divided by the time range of the day, whether it is a weekend, whether it is a holiday, where the time range of the day is divided into daytime working hours and nighttime off-duty time; the "daytime working hours" and "nighttime off-duty time" are set according to the specific conditions of different sites, and the default refers to 7 am to 7 pm and 7 pm to 7 am, where the endpoints are included before but not after.

5. The method for predicting heating load of a distributed energy station based on LSTM according to claim 4 is characterized in that: Also includes: S5-1. For any group, randomly select N groups of heating operation data as subsamples and put them into the root node of an isolated tree; S5-2. At each root node, randomly select a feature in the heating operation data, and randomly select a value between the maximum and minimum values ​​of the feature as the split value. The data is divided into the left subtree or the right subtree according to the split value; S5-3, randomly select one of the other data items of the heating operation data as a feature, and repeat step S5-2 until the data cannot be divided any further, and an isolated tree is obtained; S5-4, repeat steps S5-1 to S5-3 to obtain M isolated trees to form an isolation forest, where M is the number of isolated trees obtained by applying the isolation forest algorithm to the data in one group; S5-5. In the isolation forest, a threshold value of the number of split points is set to judge any data. During the isolation process, if the number of split points exceeds the threshold value, it is judged as normal data; if the number of split points does not exceed the threshold value, it is judged as abnormal data. S5-6. Feedback the detected abnormal data for data missing processing.

6. The method for predicting heating load of a distributed energy station based on LSTM according to claim 5 is characterized in that: The distributed energy station load prediction model includes: Three gating units, namely input gate, forget gate and output gate, are introduced to control the dimension of retaining the heating data at the previous moment. The LSTM input data includes: water supply temperature, return water temperature, instantaneous heat, instantaneous flow, indoor temperature, target room temperature, temperature, humidity, rainfall, wind speed, wind direction, weather type, solar radiation, floor, hour of the day, week of the week and whether it is a holiday. The forget gate determines the data information that LSTM discards from the heating data of the previous moment, specifically including: Set the current time to t, and the forget gate reads the output of the previous layer and current input real-time data , calculate the retention factor of the information at the previous moment : Where: is the forget gate weight matrix, is the forget gate bias, is the sigmoid function, Mapped to the interval [0,1]; where Generated by the input at the previous moment and the retained information at the previous moment; The cell state obtained from the information of the previous moment and Multiply, A value of 1 indicates that All information in the , A value of 0 means completely discarding Information in The input gate is used to screen information and determine the new heating data input at the current moment to be used to update the final output data information, including a sigmoid layer and a tanh layer; The sigmoid layer is based on the output of the previous layer. and current input real-time data Calculate the update factor of the heating information at the new time : in: is the input gate sigmoid layer weight matrix, Bias for the input gate sigmoid layer; The tanh layer calculates candidate cell states , is added to the cell state at the current moment middle: in: is the input gate tanh layer weight matrix, is the input gate tanh layer bias, is the tanh activation function; The current cell state It is equal to the sum of the product of the output value of the forget gate and the product of the two output values ​​of the input gate, that is: The output gate is used to calculate the output of the current layer ; The output gate first passes through a sigmoid layer to determine the cell state at the current moment. Output data information: Where: is the output gate weight matrix, Bias for the output gate; The current cell state After processing by tanh, a value in the interval [-1,1] is obtained and it is combined with Multiply and finally determine the output : If the current layer is the last layer, then It is the final prediction value of the distributed energy station heating load prediction model.

7. The method for predicting heating load of a distributed energy station based on LSTM according to claim 6 is characterized in that: Also includes: Set the prediction time span p and reference data time span k for the LSTM algorithm; Assume that the current time is d, and use k data from time d-k+1 to time d to predict the energy station load value at the future time p; When LSTM is required to provide the recommended load value at the current moment: in the model training stage, set p = 0, the training set is a group of k data, the target value of each group of data is the load value corresponding to the kth data, the weather data is the weather data at the corresponding moment of each data, and in the model inference stage, select the historical k-1 data and the current real-time data, combine the weather data at the corresponding moment, and apply the model to obtain the recommended load value at the current moment; When LSTM is needed to predict the recommended load value at a future time, set the future time R. In the model training phase, set p = Rd. The training set consists of k data items. The target value of each group of data is the load value corresponding to the k+pth data item. The weather data is the weather data at the time + p corresponding to each data item. In the model inference phase, select k-1 historical data items and the current real-time data, combine them with the weather data at the time + p corresponding to each data item, and apply the model to obtain the recommended load value at the future time R.

8. The method for predicting heating load of a distributed energy station based on LSTM according to claim 7 is characterized in that: The system evaluation model includes: Obtain the formed distributed energy station heating load forecasting model and the amount of missing data in the process of building the model; Define the model prediction tolerance threshold, and perform operation analysis on the formed distributed energy station heating load prediction model, wherein the operation analysis refers to: Starting from the current moment, obtain the absolute value of the difference between the actual result and the predicted result each time, and continue to accumulate until the total absolute value of the difference exceeds the model prediction tolerance threshold for the first time, and record the total number of predicted data experienced from the current moment; Taking the number of missing data as the horizontal coordinate and the total number of experienced forecast data as the vertical coordinate, linear fitting is performed on the coordinate points formed by the heating load forecasting model of all distributed energy stations to form a functional relationship between the number of missing data and the total number of experienced forecast data; Obtain the number of missing data in the current distributed energy station heating load forecasting model, feedback the total number of forecast data experienced based on the functional relationship, mark the position, and when the model reaches the marked position, feedback to the management port to remind the model to update.

9. The method for predicting heating load of a distributed energy station based on LSTM according to claim 8, characterized in that: Also includes: When a new site is put into use, it is compared with the building attribute data of the site with an existing heating load prediction model, and the sites that belong to the same group with its building attributes are calculated and the model is reused. After the new site is put into use for some time, the model is updated according to the historical operation data of the new site so that the mean room temperature of the site reaches a stable state.

Images