Method for predicting heating load of heating station

Abstract

The invention provides a heat supply load prediction method for a heating station, which belongs to the technical field of heat load prediction, and specifically comprises the following steps: determining weight values of fluctuating load users and stable load users according to the number of fluctuating load users and stable load users, the load proportion and the load fluctuation quantity of historical similar days; and the thermal load of the thermal station on the prediction day is determined by combining the corrected predicted thermal load of the fluctuating load user on the prediction day, the historical load of the stable load user on the historical similar day and the predicted thermal load of the thermal station on the prediction day, so that the prediction reliability of the thermal load of the thermal station is ensured.

Description

Technical Field

[0001] The invention belongs to the technical field of heat load prediction, and in particular to a method for predicting the heat supply load of a thermal power station. Background Art

[0002] In order to match the current heating capacity with the user's heating demand, the heat load forecast must be added to the heating system to guide the heating, so that the heating system can be pre-adjusted according to the predicted value of the heat load forecast model. Therefore, building an accurate, efficient and reliable heat load forecast model is a solid foundation for ensuring that the centralized heating system can achieve on-demand heating.

[0003] In order to realize the heat load prediction of the heating system, the invention patent CN202310566159.9 "Heat load prediction method, model training method, device and equipment for heating system" constructs a prediction feature set through weather feature data and heat load feature data, and then uses the prediction feature set and the pre-trained heat load prediction model to obtain the heat load at the target time, but there are the following technical problems:

[0004] The fluctuation of heat load in different heating areas is not the same. Specifically, due to the different proportions of industrial load and residential load in different areas, even for residential load, due to differences in building insulation, etc., the fluctuation of heat load in different heating areas is quite different. Therefore, if the accuracy of heat load prediction cannot be evaluated based on the fluctuation of heat load, the accuracy of heat load prediction may be affected to a certain extent.

[0005] In view of the above technical problems, the present invention provides a method for predicting heating load of a thermal power station. Summary of the invention

[0006] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0007] According to one aspect of the present invention, a method for predicting heating load of a thermal power station is provided.

[0008] A method for predicting heating load of a thermal power station, characterized by comprising:

[0009] S1 obtains the predicted heat load of the heat station on the predicted day and the historical similar days according to the load characteristic quantity and weather characteristic quantity of the predicted day of the heat station, and when it is determined according to the load fluctuation quantity of the historical similar days that the accuracy of the predicted heat load cannot meet the requirements, proceeds to the next step;

[0010] S2 divides the heat users of the heat station into stable load users and fluctuating load users according to the load fluctuation amount of the heat users, and divides the fluctuating load users into stable users and fluctuating users on similar days according to the load fluctuation amount on similar historical days;

[0011] S3: obtaining the predicted heat load of the fluctuating load user on the predicted day according to the load characteristic quantity and weather characteristic quantity of the predicted day, and determining the revised predicted heat load of the fluctuating load user on the predicted day in combination with the load fluctuation quantity, quantity and load proportion of the stable users and fluctuating users on similar days in history among the fluctuating load users;

[0012] S4 determines the weight values ​​of the fluctuating load users and the stable load users according to their number, load proportion and load fluctuation on similar historical days, and determines the thermal load of the thermal power station on the prediction day in combination with the revised predicted thermal load of the fluctuating load users on the prediction day, the historical load of the stable load users on similar historical days and the predicted thermal load of the thermal power station on the prediction day.

[0013] The beneficial effects of the present invention are:

[0014] By determining the accuracy of the predicted thermal load based on the load fluctuations on similar historical days, the screening of similar dates to the predicted day from the perspective of weather characteristics is achieved. At the same time, by determining the load fluctuations on similar historical days, the technical problem of low accuracy of the predicted thermal load due to excessive load fluctuations is avoided.

[0015] By classifying stable load users and fluctuating load users, and stable users and fluctuating users on similar days, the load fluctuations of different users are taken into account. At the same time, by determining the fluctuations on similar historical days, the accuracy of load forecasting is further improved.

[0016] The predicted heat load of the thermal power station is corrected by the weights of fluctuating load users, stable load users and heat load, so that the final heat load can fully take into account the fluctuations of different types of users, which not only makes the prediction result of the heat load more accurate, but also enables the heat load to be combined with the actual situation of the heat users.

[0017] A further technical solution is that the load characteristic quantities include but are not limited to historical heating loads of the same date type and historical heating loads of nearby dates whose interval with the predicted date is not greater than a set time, and the weather characteristic quantities include but are not limited to outdoor temperature, outdoor humidity, solar radiation and wind speed.

[0018] A further technical solution is that the historical similar day is determined according to weather characteristic quantities on different dates, specifically according to deviations of weather characteristic quantities on different dates and correlations between different weather characteristic quantities and heating load.

[0019] A further technical solution is that the specific steps of the historical similar day evaluation are:

[0020] Using principal component analysis to determine correlation coefficients between the different weather characteristic quantities and the heating load, and determining weight factors of the different weather characteristic quantities based on the correlation coefficients between the different weather characteristic quantities and the heating load;

[0021] Dividing the weather characteristic quantities into strongly correlated weather characteristic quantities and weakly correlated weather characteristic quantities according to the weight factors of the different weather characteristic quantities;

[0022] Screening historical days by the similar number of the weather characteristic quantities to obtain candidate historical days, and further screening the candidate similar days by the similar number of the strongly correlated weather characteristic quantities to obtain suspected historical similar days;

[0023] The similarity between the suspected historical similar day and the predicted day is determined by the number of similar strongly correlated weather feature quantities and weight factors and the number of similar weakly correlated weather feature quantities and weight factors of the suspected historical similar day and the predicted day, and the historical similar day is determined based on the similarity.

[0024] A further technical solution is to determine that the accuracy of the predicted heat load cannot meet the requirements according to the load fluctuation amount of the historical similar day, specifically including:

[0025] S11 determines the benchmark load of the historically similar day according to the average value of the load of the historically similar day, and determines the load fluctuation amount of the historically similar day according to the deviation amount between the load of the historically similar day and the benchmark load;

[0026] S12 determines whether the maximum value of the load fluctuation amount of the historical similar day is greater than the preset maximum deviation threshold value. If so, it is determined that the accuracy of the predicted thermal load cannot meet the requirements. If not, proceed to the next step;

[0027] S13 takes the historical similar days with load fluctuations greater than the preset deviation threshold as deviation similar days, and determines whether the accuracy of the predicted heat load on the predicted day fails to meet the requirements according to the number of the deviation similar days and their proportion in the historical similar days. If so, proceed to the next step; if not, determine that the accuracy of the predicted heat load fails to meet the requirements;

[0028] S14 determines the estimated accuracy of the predicted thermal load by the number of the deviation similar days and the average of the load fluctuation, the proportion in the historical similar days, and the average of the load fluctuation of the historical similar days, and determines whether the accuracy of the measured thermal load fails to meet the requirements through the estimated accuracy.

[0029] A further technical solution is to divide the fluctuating load users into similar day stable users and fluctuating users according to the load fluctuation amount on similar historical days, specifically including:

[0030] When the maximum value of the load fluctuation amount of the fluctuating load user on the historical similar day is less than a preset threshold, the fluctuating load user is classified as a stable user on the similar day;

[0031] When the maximum value of the load fluctuation amount of the fluctuating load user on the historical similar day is not less than the preset threshold, the fluctuating load user is divided into similar day stable users and fluctuating users according to the proportion of historical similar days with the load fluctuation amount greater than the preset deviation threshold.

[0032] A further technical solution is that the method for determining the corrected predicted heat load of the fluctuating load user on the predicted day is:

[0033] S31 determines the stable corrected predicted thermal load of the fluctuating load user by combining the load fluctuation amount of the stable users on similar days in history among the fluctuating load users, the number of stable users on similar days among the fluctuating load users, and the load proportion among the fluctuating load users;

[0034] S32 determines the reference value of the modified predicted thermal load of the fluctuating load user through the stable modified predicted thermal load and the predicted thermal load of the fluctuating load user, and determines whether the reference value of the modified predicted thermal load of the fluctuating load user is accurate through the number of fluctuating load users and the number ratio, the load and the load ratio among the fluctuating load users, if so, proceeds to the next step, if not, proceeds to step S34;

[0035] S33 calculates the maximum value and the average value of the load fluctuation amount of the fluctuating users among the fluctuating load users, and determines whether the reference value of the corrected predicted thermal load of the fluctuating load users is accurate based on the maximum value and the average value of the load fluctuation amount of the fluctuating users. If so, the corrected predicted thermal load of the fluctuating load users is determined by the reference value of the corrected predicted thermal load of the fluctuating load users. If not, proceed to step S34.

[0036] S34 determines the load correction amount of the fluctuating load user through the maximum value and average value of the load fluctuation amount, the number and the proportion of the number, the load and the load proportion of the fluctuating load user, and determines the corrected predicted thermal load of the fluctuating load user in combination with the reference value of the corrected predicted thermal load of the fluctuating load user.

[0037] A further technical solution is to determine whether the baseline value of the corrected predicted thermal load of the fluctuating load users is accurate by determining the number of fluctuating load users among the fluctuating load users and the load when and only when the proportion of the number of fluctuating load users and the load proportion among the fluctuating load users are both smaller than the predicted proportion.

[0038] A further technical solution is that the method for determining the weight value of the fluctuating load user is:

[0039] Determine the benchmark weight value of the fluctuating load user by the proportion of the number of heat users in the thermal power station and the proportion of the load of the fluctuating load user;

[0040] Determine the fluctuation weight value of the fluctuating load user by normalizing the average value of the load fluctuation amount of the fluctuating load user on similar historical days with the average value of the load fluctuation amount of the stable load user on similar historical days;

[0041] The weight value of the fluctuating load user is determined by the fluctuating weight value of the fluctuating load user and the reference weight value.

[0042] A further technical solution is that the method for determining the heat load of the thermal power station on the forecast day is:

[0043] Determine the estimated thermal load of the thermal power station on the forecast day according to the weight value of the fluctuating load user and the corrected forecast thermal load of the fluctuating load user on the forecast day, the weight value of the stable load user and the average of the historical load of the stable load user on similar historical days;

[0044] When the deviation between the assessed heat load of the thermal power station on the forecast day and the forecast heat load is less than the forecast load deviation:

[0045] Determining the heat load of the heat power station on the forecast day by taking the average value of the assessed heat load and the forecast heat load of the heat power station on the forecast day;

[0046] When the deviation between the assessed heat load of the thermal power station on the forecast day and the forecast heat load is not less than the forecast load deviation:

[0047] Determining the similarity between the predicted day and the historical similar day through the weather characteristic quantity of the predicted day, and determining the consistent historical day through the similarity of the historical similar day;

[0048] The accuracy of the evaluated heat load is determined by the similarity and number of the historical similar days and the similarity and number of the consistent historical days. When the accuracy of the evaluated heat load is greater than a preset accuracy threshold, the heat load of the thermal power station on the predicted day is determined by the evaluated heat load of the thermal power station on the predicted day. When the accuracy of the evaluated heat load is not greater than the preset accuracy threshold, the heat load of the thermal power station on the predicted day is determined by the predicted heat load of the thermal power station on the predicted day.

[0049] On the other hand, a computer-readable storage medium is provided in an embodiment of the present application, on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the above-mentioned method for predicting the heating load of a thermal power station.

[0050] On the other hand, a computer program product is provided in an embodiment of the present application, characterized in that the computer program product stores instructions, and when the instructions are executed by a computer, the computer implements the above-mentioned method for predicting heating load of a thermal power station. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0052] Figure 1 This is a flow chart of a method for predicting heating load of a thermal power station according to Example 1.

[0053] Figure 2 It is a flowchart of specific steps of historical similar day evaluation according to Example 1.

[0054] Figure 3 It is a framework diagram of a computer-readable storage medium according to Example 2. DETAILED DESCRIPTION

[0055] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of ways and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0056] The terms "a", "an", "the", and "said" are used to indicate the presence of one or more elements / components / etc.; the terms "comprising" and "having" are used to express an open-ended inclusive meaning and mean that additional elements / components / etc. may be present in addition to the listed elements / components / etc.

[0057] Example 1

[0058] To solve the above problems, according to one aspect of the present invention, Figure 1 As shown, a method for predicting heating load of a thermal power station is provided, which is characterized by specifically comprising:

[0059] S1 obtains the predicted heat load of the heat station on the predicted day and the historical similar days according to the load characteristic quantity and weather characteristic quantity of the predicted day of the heat station, and when it is determined according to the load fluctuation quantity of the historical similar days that the accuracy of the predicted heat load cannot meet the requirements, proceeds to the next step;

[0060] Furthermore, the load characteristic quantities include but are not limited to historical heating loads of the same date type and historical heating loads of nearby dates whose interval with the predicted date is not greater than a set time, and the weather characteristic quantities include but are not limited to outdoor temperature, outdoor humidity, solar radiation and wind speed.

[0061] Furthermore, the historical similar day is determined according to weather characteristic quantities on different dates, and specifically determined according to deviations of weather characteristic quantities on different dates and correlations between different weather characteristic quantities and heating loads.

[0062] Further, such as Figure 2 As shown, the specific steps of the historical similar day evaluation are:

[0063] Using principal component analysis to determine correlation coefficients between the different weather characteristic quantities and the heating load, and determining weight factors of the different weather characteristic quantities based on the correlation coefficients between the different weather characteristic quantities and the heating load;

[0064] Dividing the weather characteristic quantities into strongly correlated weather characteristic quantities and weakly correlated weather characteristic quantities according to the weight factors of the different weather characteristic quantities;

[0065] Screening historical days by the similar number of the weather characteristic quantities to obtain candidate historical days, and further screening the candidate similar days by the similar number of the strongly correlated weather characteristic quantities to obtain suspected historical similar days;

[0066] The similarity between the suspected historical similar day and the predicted day is determined by the number of similar strongly correlated weather feature quantities and weight factors and the number of similar weakly correlated weather feature quantities and weight factors of the suspected historical similar day and the predicted day, and the historical similar day is determined based on the similarity.

[0067] In one possible embodiment, determining that the accuracy of the predicted heat load cannot meet the requirement based on the load fluctuation amount of the historical similar day specifically includes:

[0068] S11 determines the benchmark load of the historically similar day according to the average value of the load of the historically similar day, and determines the load fluctuation amount of the historically similar day according to the deviation amount between the load of the historically similar day and the benchmark load;

[0069] S12 determines whether the maximum value of the load fluctuation amount of the historical similar day is greater than the preset maximum deviation threshold value. If so, it is determined that the accuracy of the predicted thermal load cannot meet the requirements. If not, proceed to the next step;

[0070] S13 takes the historical similar days with load fluctuations greater than the preset deviation threshold as deviation similar days, and determines whether the accuracy of the predicted heat load on the predicted day fails to meet the requirements according to the number of the deviation similar days and their proportion in the historical similar days. If so, proceed to the next step; if not, determine that the accuracy of the predicted heat load fails to meet the requirements;

[0071] S14 determines the estimated accuracy of the predicted thermal load by the number of the deviation similar days and the average of the load fluctuation, the proportion in the historical similar days, and the average of the load fluctuation of the historical similar days, and determines whether the accuracy of the measured thermal load fails to meet the requirements through the estimated accuracy.

[0072] By determining the accuracy of the predicted thermal load based on the load fluctuations on similar historical days, the screening of similar dates to the predicted day from the perspective of weather characteristics is achieved. At the same time, by determining the load fluctuations on similar historical days, the technical problem of low accuracy of the predicted thermal load due to excessive load fluctuations is avoided.

[0073] S2 divides the heat users of the heat station into stable load users and fluctuating load users according to the load fluctuation amount of the heat users, and divides the fluctuating load users into stable users and fluctuating users on similar days according to the load fluctuation amount on similar historical days;

[0074] Specifically, the fluctuating load users are divided into similar day stable users and fluctuating users according to the load fluctuation amount on similar historical days, specifically including:

[0075] When the maximum value of the load fluctuation amount of the fluctuating load user on the historical similar day is less than a preset threshold, the fluctuating load user is classified as a stable user on the similar day;

[0076] When the maximum value of the load fluctuation amount of the fluctuating load user on the historical similar day is not less than the preset threshold, the fluctuating load user is divided into similar day stable users and fluctuating users according to the proportion of historical similar days with the load fluctuation amount greater than the preset deviation threshold.

[0077] By classifying stable load users and fluctuating load users, and stable users and fluctuating users on similar days, the load fluctuations of different users are taken into account. At the same time, by determining the fluctuations on similar historical days, the accuracy of load forecasting is further improved.

[0078] S3: obtaining the predicted heat load of the fluctuating load user on the predicted day according to the load characteristic quantity and weather characteristic quantity of the predicted day, and determining the revised predicted heat load of the fluctuating load user on the predicted day in combination with the load fluctuation quantity, quantity and load proportion of the stable users and fluctuating users on similar days in history among the fluctuating load users;

[0079] Furthermore, the method for determining the corrected predicted heat load of the fluctuating load user on the predicted day is:

[0080] S31 determines the stable corrected predicted thermal load of the fluctuating load user by combining the load fluctuation amount of the stable users on similar days in history among the fluctuating load users, the number of stable users on similar days among the fluctuating load users, and the load proportion among the fluctuating load users;

[0081] S32 determines the reference value of the modified predicted thermal load of the fluctuating load user through the stable modified predicted thermal load and the predicted thermal load of the fluctuating load user, and determines whether the reference value of the modified predicted thermal load of the fluctuating load user is accurate through the number of fluctuating load users and the number ratio, the load and the load ratio among the fluctuating load users, if so, proceeds to the next step, if not, proceeds to step S34;

[0082] S33 calculates the maximum value and the average value of the load fluctuation amount of the fluctuating users among the fluctuating load users, and determines whether the reference value of the corrected predicted thermal load of the fluctuating load users is accurate based on the maximum value and the average value of the load fluctuation amount of the fluctuating users. If so, the corrected predicted thermal load of the fluctuating load users is determined by the reference value of the corrected predicted thermal load of the fluctuating load users. If not, proceed to step S34.

[0083] S34 determines the load correction amount of the fluctuating load user through the maximum value and average value of the load fluctuation amount, the number and the proportion of the number, the load and the load proportion of the fluctuating load user, and determines the corrected predicted thermal load of the fluctuating load user in combination with the reference value of the corrected predicted thermal load of the fluctuating load user.

[0084] Specifically, when and only when the proportion of the number of fluctuating load users and the proportion of load among the fluctuating load users are both smaller than the predicted proportion, the accuracy of the baseline value of the corrected predicted thermal load of the fluctuating load users is determined by determining the number of fluctuating load users among the fluctuating load users and the load.

[0085] S4 determines the weight values ​​of the fluctuating load users and the stable load users according to their number, load proportion and load fluctuation on similar historical days, and determines the thermal load of the thermal power station on the prediction day in combination with the revised predicted thermal load of the fluctuating load users on the prediction day, the historical load of the stable load users on similar historical days and the predicted thermal load of the thermal power station on the prediction day.

[0086] Optionally, when predicting the heat load of the thermal power station on the forecast day, a prediction model based on the HHO-SA algorithm is constructed, specifically including:

[0087] (1) Divide the preprocessed dataset into training set and test set.

[0088] (2) Initialize the parameters h, n, l of the GRU-SA model, and use the value range of h, n, l of IHHO as the optimization space of IHHO. Use the best point set method to generate a population of N, and set the maximum number of iterations T.

[0089] (3) Input the training set data into the GRU-SA model for training, and use the root mean square error (RMSE) generated by the trained model on the test set as the fitness objective function.

[0090] (4) Calculate the fitness value, increase the number of iterations by one, and select a specific strategy to update the position of the population according to the value of the prey escape energy E and the random parameter r in IHHO, that is, the value of the parameters h, n, and l.

[0091] (5) When the number of iterations reaches the maximum number of iterations, the parameter optimization ends. If the fitness value converges, the optimal GRU-SA model parameter combination is obtained; otherwise, return to step 3.

[0092] (6) Input the test set data into the trained optimal GRU-SA model to obtain the predicted value of the day to be predicted.

[0093] Optionally, in order to improve the model prediction performance of GRU-SA, a prediction model based on HHO-GRU-SA is built based on GRU-SA and combined with the HHO algorithm. This model uses the root mean square error RMSE predicted by GRU-SA as the fitness function, that is, the dependent variable; the number of hidden layers h, the number of hidden layer neurons n and the learning rate l as the independent variables to be optimized, and the formula is expressed as follows:

[0094] Y RMSE =f(h,n,l)

[0095] This formula combines the unique global search and local search position update strategy of IHHO, and finds the optimal solution to the problem after a set maximum number of iterative optimizations. If the algorithm convergence parameter combination is h, n, l, then the optimal parameter of GRU-SA is [h best , n best , l best ], at this time, the RMSEY value of the model is also the smallest and the prediction accuracy is also the highest.

[0096] Furthermore, the method for determining the weight value of the fluctuating load user is:

[0097] Determine the benchmark weight value of the fluctuating load user by the proportion of the number of heat users in the thermal power station and the proportion of the load of the fluctuating load user;

[0098] Determine the fluctuation weight value of the fluctuating load user by normalizing the average value of the load fluctuation amount of the fluctuating load user on similar historical days with the average value of the load fluctuation amount of the stable load user on similar historical days;

[0099] The weight value of the fluctuating load user is determined by the fluctuating weight value of the fluctuating load user and the reference weight value.

[0100] Specifically, the method for determining the heat load of the thermal power station on the forecast day is:

[0101] Determine the estimated thermal load of the thermal power station on the forecast day according to the weight value of the fluctuating load user and the corrected forecast thermal load of the fluctuating load user on the forecast day, the weight value of the stable load user and the average of the historical load of the stable load user on similar historical days;

[0102] When the deviation between the assessed heat load of the thermal power station on the forecast day and the forecast heat load is less than the forecast load deviation:

[0103] Determining the heat load of the heat power station on the forecast day by taking the average value of the assessed heat load and the forecast heat load of the heat power station on the forecast day;

[0104] When the deviation between the assessed heat load of the thermal power station on the forecast day and the forecast heat load is not less than the forecast load deviation:

[0105] Determining the similarity between the predicted day and the historical similar day through the weather characteristic quantity of the predicted day, and determining the consistent historical day through the similarity of the historical similar day;

[0106] The accuracy of the evaluated heat load is determined by the similarity and number of the historical similar days and the similarity and number of the consistent historical days. When the accuracy of the evaluated heat load is greater than a preset accuracy threshold, the heat load of the thermal power station on the predicted day is determined by the evaluated heat load of the thermal power station on the predicted day. When the accuracy of the evaluated heat load is not greater than the preset accuracy threshold, the heat load of the thermal power station on the predicted day is determined by the predicted heat load of the thermal power station on the predicted day.

[0107] The predicted heat load of the thermal power station is corrected by the weights of fluctuating load users, stable load users and heat load, so that the final heat load can fully take into account the fluctuations of different types of users, which not only makes the prediction result of the heat load more accurate, but also enables the heat load to be combined with the actual situation of the heat users.

[0108] Example 2

[0109] like Figure 3 As shown, in an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute the above-mentioned method for predicting the heating load of a thermal power station.

[0110] Example 3

[0111] A computer program product is provided in an embodiment of the present application, characterized in that the computer program product stores instructions, and when the instructions are executed by a computer, the computer implements the above-mentioned method for predicting heating load of a thermal power station.

[0112] In the embodiments of the present invention, the term "plurality" refers to two or more than two, unless otherwise clearly defined. The terms "installation", "connection", "fixation" and the like should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection. For those skilled in the art, the specific meanings of the above terms in the embodiments of the present invention can be understood according to the specific circumstances.

[0113] In the description of the embodiments of the present invention, it needs to be understood that the directions or positional relationships indicated by the terms "upper" and "lower" are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, rather than indicating or implying that the device or unit referred to must have a specific direction, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the embodiments of the present invention.

[0114] In the description of this specification, the description of the terms "one embodiment", "a preferred embodiment", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0115] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting heating load of a thermal power station, characterized in that: Specifically include: According to the load characteristic quantity and weather characteristic quantity of the thermal power station on the predicted day, the predicted thermal load of the thermal power station on the predicted day and the historical similar days are obtained, and when it is determined according to the load fluctuation quantity of the historical similar days that the accuracy of the predicted thermal load cannot meet the requirements, proceed to the next step; According to the load fluctuation amount of the heat users of the heat station, the heat users are divided into stable load users and fluctuating load users, and according to the load fluctuation amount of similar historical days, the fluctuating load users are divided into stable users and fluctuating users on similar days; The predicted thermal load of the fluctuating load user on the predicted day is obtained according to the load characteristic quantity and weather characteristic quantity of the predicted day, and the corrected predicted thermal load of the fluctuating load user on the predicted day is determined by combining the load fluctuation quantity, quantity and load proportion of similar-day stable users and fluctuating users among the fluctuating load users; The weight values ​​of the fluctuating load users and the stable load users are determined according to their number, load proportion and load fluctuation on similar historical days, respectively; and the thermal load of the thermal power station on the prediction day is determined in combination with the revised predicted thermal load of the fluctuating load users on the prediction day, the historical load of the stable load users on similar historical days and the predicted thermal load of the thermal power station on the prediction day.

2. The method for predicting heating load of a thermal power station according to claim 1, characterized in that: The load characteristic quantities include but are not limited to historical heating loads of the same date type and historical heating loads of nearby dates whose interval with the forecast date is not greater than a set time. The weather characteristic quantities include but are not limited to outdoor temperature, outdoor humidity, solar radiation and wind speed.

3. The method for predicting heating load of a thermal power station according to claim 1, characterized in that: The historical similar day is determined according to weather characteristic quantities on different dates, and is specifically determined according to deviations of weather characteristic quantities on different dates and correlations between different weather characteristic quantities and heating loads.

4. The method for predicting heating load of a thermal power station according to claim 3, characterized in that: The specific steps of the historical similar day evaluation are: Using principal component analysis to determine correlation coefficients between the different weather characteristic quantities and the heating load, and determining weight factors of the different weather characteristic quantities based on the correlation coefficients between the different weather characteristic quantities and the heating load; Dividing the weather characteristic quantities into strongly correlated weather characteristic quantities and weakly correlated weather characteristic quantities according to the weight factors of the different weather characteristic quantities; Screening historical days by the similar number of the weather characteristic quantities to obtain candidate historical days, and further screening the candidate similar days by the similar number of the strongly correlated weather characteristic quantities to obtain suspected historical similar days; The similarity between the suspected historical similar day and the predicted day is determined by the number of similar strongly correlated weather feature quantities and weight factors and the number of similar weakly correlated weather feature quantities and weight factors of the suspected historical similar day and the predicted day, and the historical similar day is determined based on the similarity.

5. The method for predicting heating load of a thermal power station according to claim 1, characterized in that: The accuracy of the predicted heat load determined based on the load fluctuation of the historical similar day cannot meet the requirements, specifically including: Determine the benchmark load of the historically similar day according to the average value of the load of the historically similar day, and determine the load fluctuation amount of the historically similar day according to the deviation amount between the load of the historically similar day and the benchmark load; Determine whether the maximum value of the load fluctuation amount of the historical similar day is greater than the preset maximum deviation threshold value, if so, determine that the accuracy of the predicted thermal load cannot meet the requirements, if not, proceed to the next step; The historical similar days with a load fluctuation greater than a preset deviation threshold are regarded as deviation similar days, and according to the number of the deviation similar days and their proportion in the historical similar days, it is determined whether the accuracy of the predicted heat load on the predicted day cannot meet the requirements, if so, proceed to the next step, if not, it is determined that the accuracy of the predicted heat load cannot meet the requirements; The estimated accuracy of the predicted heat load is determined by the number of the deviation similar days and the average of the load fluctuations, their proportion in the historical similar days, and the average of the load fluctuations in the historical similar days, and the estimated accuracy is used to determine whether the accuracy of the measured heat load fails to meet the requirements.

6. The method for predicting heating load of a thermal power station according to claim 1, characterized in that: According to the load fluctuation amount on similar historical days, the fluctuating load users are divided into similar day stable users and fluctuating users, specifically including: When the maximum value of the load fluctuation amount of the fluctuating load user on the historical similar day is less than a preset threshold, the fluctuating load user is classified as a stable user on the similar day; When the maximum value of the load fluctuation amount of the fluctuating load user on the historical similar day is not less than the preset threshold, the fluctuating load user is divided into similar day stable users and fluctuating users according to the proportion of historical similar days with the load fluctuation amount greater than the preset deviation threshold.

7. The method for predicting heating load of a thermal power station according to claim 1, characterized in that: The method for determining the weight value of the fluctuating load user is: Determine the benchmark weight value of the fluctuating load user by the proportion of the number of heat users in the thermal power station and the proportion of the load of the fluctuating load user; Determine the fluctuation weight value of the fluctuating load user by normalizing the average value of the load fluctuation amount of the fluctuating load user on similar historical days with the average value of the load fluctuation amount of the stable load user on similar historical days; The weight value of the fluctuating load user is determined by the fluctuating weight value of the fluctuating load user and the reference weight value.

8. The method for predicting heating load of a thermal power station according to claim 1, characterized in that: The method for determining the heat load of the thermal power station on the forecast day is: Determine the estimated thermal load of the thermal power station on the forecast day according to the weight value of the fluctuating load user and the corrected forecast thermal load of the fluctuating load user on the forecast day, the weight value of the stable load user and the average of the historical load of the stable load user on similar historical days; When the deviation between the assessed heat load and the predicted heat load of the thermal power station on the predicted day is less than the predicted load deviation: Determining the heat load of the heat power station on the forecast day by taking the average value of the assessed heat load and the forecast heat load of the heat power station on the forecast day; When the deviation between the assessed heat load of the thermal power station on the forecast day and the forecast heat load is not less than the forecast load deviation: Determining the similarity between the predicted day and the historical similar day through the weather characteristic quantity of the predicted day, and determining the consistent historical day through the similarity of the historical similar day; The accuracy of the evaluated heat load is determined by the similarity and number of the historical similar days and the similarity and number of the consistent historical days. When the accuracy of the evaluated heat load is greater than a preset accuracy threshold, the heat load of the thermal power station on the predicted day is determined by the evaluated heat load of the thermal power station on the predicted day. When the accuracy of the evaluated heat load is not greater than the preset accuracy threshold, the heat load of the thermal power station on the predicted day is determined by the predicted heat load of the thermal power station on the predicted day.

9. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute a method for predicting heating load of a thermal power station as claimed in any one of claims 1 to 8.

10. A computer program product, characterized in that The computer program product stores instructions, and when the instructions are executed by a computer, the computer implements a method for predicting heating load of a thermal power station as described in any one of claims 1-8.

Images