A multi-energy complementary new energy power plant intelligent operation and maintenance method, system and device

Abstract

The application relates to the technical field of new energy power station operation and maintenance, and discloses a multi-energy complementary new energy power station intelligent operation and maintenance method, system and equipment, which comprises the following steps: based on parameter data and corresponding meteorological data of multiple new energy power stations, establishing a power generation capacity prediction model of each new energy power station to preset power generation capacity; analyzing various types of unit operation related data of multiple new energy power stations collected by SCADA systems and CMS systems to obtain analysis data; based on the power generation capacity prediction model and the analysis data of each new energy power station, establishing a fault model and an efficiency evaluation model of each type of unit for fault diagnosis and efficiency evaluation; and based on the power generation capacity predicted by the power generation capacity prediction model of each new energy power station, determining the power generation configuration of each new energy power station. The application can monitor the fault and evaluate the efficiency of the operation state of each new energy power station by establishing multiple models, and can reasonably configure the power generation configuration of each power station, thereby improving the operation and maintenance efficiency and reliability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy power station operation and maintenance, in particular to a multi-energy complementary new energy power station intelligent operation and maintenance method, system and equipment. BACKGROUND

[0002] New energy power station operation and maintenance lasts from the completion of power station and grid connection to the end of power station life cycle, usually lasting for ten to dozens of years, which is an important link to ensure normal operation, play the potential of installed capacity, meet electricity demand and create economic benefits. New energy power station operation and maintenance has been fully electrified, automated and networked after years of development, and has entered the intelligent era with the continuous breakthroughs of big data and artificial intelligence. The power station cluster is huge and contains various energy forms such as wind power, photovoltaic and energy storage, but there is a lack of mature operation and maintenance experience for such power stations worldwide.

[0003] Traditional new energy power station operation and maintenance still has great deficiencies in information dimension, accuracy and reliability. Modern wind turbines and photovoltaic components face many challenges in operation reliability: high-power wind energy absorption and frequent frequency modulation demand of new energy power grid system form complex wind grid load excitation to wind turbine transmission system; the manufacturing and installation cost of key components of wind turbines is high, and the downtime loss caused by faults is huge; photovoltaic components operate in harsh environments, and faults are difficult to detect, resulting in low operation and maintenance efficiency and poor reliability. SUMMARY

[0004] Therefore, in order to solve the defects of low monitoring efficiency and poor reliability of new energy power station operation and maintenance in the prior art, the present application provides a multi-energy complementary new energy power station intelligent operation and maintenance method, system and equipment.

[0005] In a first aspect, the present application provides a multi-energy complementary new energy power station intelligent operation and maintenance method, comprising:

[0006] Based on the parameter data and corresponding meteorological data of a plurality of new energy power stations, a power generation prediction model of each new energy power station is established to preset the power generation;

[0007] The analysis data is obtained by analyzing various types of unit operation related data of a plurality of new energy power stations collected by the SCADA system and the CMS system;

[0008] Based on the power generation prediction model and the analysis data of each new energy power station, a fault model and an efficiency evaluation model of each type of unit are established for fault diagnosis and efficiency evaluation;

[0009] The power generation configuration of each new energy power station is determined based on the power generation predicted by the power generation prediction model of each new energy power station.

[0010] The intelligent operation and maintenance method for multi-energy complementary new energy power plants provided in this embodiment can monitor the operating status of multiple new energy power plants in real time, perform fault diagnosis and performance evaluation, and determine the power generation configuration of each new energy power plant based on the measured power generation, which greatly improves the operation and maintenance efficiency and reliability of new energy power plants.

[0011] In one optional implementation, multiple new energy power plants include: wind energy storage power plants and solar energy storage power plants. The power generation prediction models for each new energy power plant include: a wind resource power generation prediction model and a solar resource power generation prediction model, wherein:

[0012] The wind power generation prediction model is based on the numerical simulation model of the flow field of the wind power base and the distribution of the turbines. It obtains the velocity and turbulence distribution in the wind farm to predict the power generation of the wind turbines.

[0013] The photovoltaic power generation prediction model is based on the digital twin model of photovoltaic power station equipment. It analyzes the mechanism of photovoltaic modules, the influence of clouds and solar irradiance, and obtains the influence of clouds on solar irradiance distribution to predict the power generation of photovoltaic modules.

[0014] In one alternative implementation, the process based on a wind resource power generation prediction model includes:

[0015] According to different arrangement methods of wind farms w Wheel hub height h w Wind turbine diameter d w Establish a numerical simulation model F of the flow field in the wind power base:

[0016] F=∑(h w ,d w ,l w )

[0017] Based on the numerical simulation model of the flow field in the wind power base, the velocity and turbulence intensity distribution in the wind farm are obtained by using the large eddy simulation method. The average flow velocity of the wind turbine accessories is obtained based on the velocity and turbulence intensity distribution in the wind farm.

[0018] According to different fan arrangement methods w Average flow velocity v of fan accessories w and wind energy conversion rate δ w Establish a wind power generation prediction model to predict wind farm power generation W w :

[0019] W w =∑(δ) w ,v w ,l w ).

[0020] The embodiment of the application can deeply understand the wake effect of the large wind power base through the wind power base flow field fine numerical simulation method, quickly obtain the velocity and turbulence intensity distribution in the wind power field in combination with the wind turbine characteristic parameters, and then accurately predict the wind power generation capacity through the wind resource generation capacity prediction model established in combination with the wind energy conversion rate of different arrangement modes of the wind turbine.

[0021] In an alternative embodiment, the process of predicting the power generation capacity of the photovoltaic module based on the light resource evaluation power prediction model comprises:

[0022] A physical mechanism model of the photovoltaic module is constructed:

[0023]

[0024] wherein J(V) is the current density, J sc is the short circuit current, J0 is the reverse saturation current, A is the cross-sectional area of the solar cell, Rs is the series resistance, R SH is the parallel resistance, J is the current surface density, q is the charge unit, V is the voltage of the photovoltaic module, k B is the solar panel temperature coefficient, T a is the ambient temperature;

[0025] The cloud height and cloud thickness are measured by using a sky imager, and a curve of the relationship between the cloud thickness and the cloud height and the cloud movement trajectory over time is fitted;

[0026] The solar irradiance is measured by using an irradiance table, and a camera response function of the mapping relationship between the image gray value and the brightness value received by the camera is established according to the cloud influence, and the natural logarithm of both sides is taken to obtain

[0027] g(Z i j )=ln I i +ln t j

[0028] wherein Z is the picture pixel value, I is the scene brightness, Z and t are known, and I is unknown, under the assumption that the scene brightness is constant, E is in a proportional relationship with t, for the i-th pixel, let I i =1, then t j =E i , E is the brightness received by the camera, and the response curve is fitted;

[0029] The photovoltaic module characteristics and the irradiance conversion rate δ s of the photovoltaic module represented by the physical mechanism model of the photovoltaic module are used to establish a light resource generation capacity prediction model to predict the power generation capacity W s of the photovoltaic power station:

[0030] W s =Σ(δ sJ(V).

[0031] The present application analyzes the influence of photovoltaic module mechanism, cloud cluster and solar radiation based on photovoltaic field station equipment digital twin modeling technology research, and the power generation of the light resource can be accurately predicted based on the power generation prediction model of the photovoltaic module characteristics and the irradiance conversion rate.

[0032] In an optional embodiment, the process of analyzing the various types of unit operation related data collected by the SCADA system and the CMS system of the new energy power station includes:

[0033] The various unit operation related data in the SCADA system and the CMS system are divided into multiple time series according to a preset time resolution;

[0034] The similarity between the electrical data sequence of the unit measured based on the Pearson correlation coefficient and the electrical time sequence obtained by the upscaling method is calculated to obtain a similarity calculation result;

[0035] The deviation degree between the electrical data sequence of the unit measured and the electrical time sequence obtained by the upscaling method based on the Euclidean distance is calculated to obtain a deviation degree calculation result;

[0036] The key parameters of different unit operation states are extracted based on the similarity calculation result and the deviation degree result.

[0037] Based on the similarity calculation result and the deviation degree result, the key parameters affecting the operation state of each unit of the new energy power station are deeply mined, a key influence parameter qualitative classification method based on the influence mode is researched, and the influence mechanism and coupling relationship of different levels of key influence parameters on the operation state of the wind turbine are revealed.

[0038] In an optional embodiment, the process of establishing a fault model and an efficiency evaluation model for each type of unit based on the power generation prediction model of each new energy power station and the analysis data includes:

[0039] Based on the key parameters of the wind turbine and photovoltaic module operation state, the migration component analysis algorithm is used to distribute and assimilate the operation data of different wind turbines and photovoltaic modules;

[0040] Based on the normal behavior model of the wind turbine and photovoltaic module, the key influence parameter feature vector after distribution and assimilation under all operation states is input to obtain the model output power;

[0041] The random forest algorithm is used to predict the residual error evaluation index based on the model output power to divide the different operation states of the wind turbine and photovoltaic module;

[0042] According to the running state, the fault condition of a single unit is acquired, the overall energy efficiency of the wind turbine and the photovoltaic component is obtained through cumulative summation, and the energy efficiency of the single unit is evaluated through time scale change.

[0043] The embodiment of the present application can effectively monitor the fault state and evaluate the energy efficiency of each energy power station by establishing the fault model and the efficiency evaluation model, thereby improving the operation and maintenance efficiency.

[0044] In an optional embodiment, the process of determining the power generation configuration of each new energy power station based on the predicted power generation of the power generation prediction model of each new energy power station comprises:

[0045] The meteorological data of the wind energy storage power station and the light energy storage power station in a preset historical time period is acquired and input into the wind resource power generation prediction model and the light resource power generation prediction model to obtain the corresponding power generation;

[0046] The power generation configuration of each new energy power station is determined by constructing an optimization configuration model with the maximum annual revenue of the source network load storage life cycle as the objective function and the power balance, the output of the new energy power station and the local consumption of the new energy power station as the boundary conditions, considering the reduction of the load power consumption cost and the corresponding grid access income of the power generation of each new energy power station.

[0047] The embodiment of the present application can realize energy complementation between multiple energy power stations and make full and effective use of electric energy by optimizing the configuration model to ensure the coordinated operation of the multiple new energy power station system.

[0048] In a second aspect, the present application provides a multi-energy complementary new energy power station intelligent operation and maintenance system, which comprises:

[0049] The power generation prediction model establishment module is used to establish the power generation prediction model of each new energy power station based on the parameter data and corresponding meteorological data of multiple new energy power stations to preset the power generation;

[0050] The operation data analysis module is used to analyze the various types of unit operation related data of multiple new energy power stations collected by the SCADA system and the CMS system to obtain analysis data;

[0051] The fault model and efficiency evaluation model establishment module is used to establish the fault model and efficiency evaluation model of each type of unit based on the power generation prediction model and the analysis data of each new energy power station, which is used for fault diagnosis and efficiency evaluation;

[0052] The power generation configuration module is used to determine the power generation configuration of each new energy power station based on the predicted power generation of the power generation prediction model of each new energy power station.

[0053] In an optional embodiment, the operation data analysis module comprises:

[0054] a time series division unit configured to divide the operation-related data of each unit in the SCADA system and the CMS system into a plurality of time series according to a preset time resolution;

[0055] a similarity calculation unit configured to calculate a similarity between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method based on a Pearson correlation coefficient, to obtain a similarity calculation result;

[0056] a deviation degree calculation unit configured to calculate a deviation degree between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method based on a Euclidean distance, to obtain a deviation degree calculation result;

[0057] a key parameter extraction unit configured to extract key parameters of different unit operation states based on the similarity calculation result and the deviation degree result.

[0058] In an optional embodiment, the fault model and performance evaluation model establishment module comprises:

[0059] an operation data distribution assimilation unit configured to perform operation data distribution assimilation of different wind turbines and photovoltaic modules based on the key parameters of the operation states of the wind turbines and the photovoltaic modules by using a transfer component analysis algorithm;

[0060] an output power acquisition unit configured to acquire a model based on the normal behavior model of the wind turbines and the photovoltaic modules, with the distribution-assimilated key influence parameter feature vectors in all operation states as input;

[0061] an operation state division unit configured to divide different operation states of the wind turbines and the photovoltaic modules based on a model output power prediction residual error evaluation index by using a random forest algorithm;

[0062] an energy efficiency evaluation unit configured to obtain a fault condition of a single unit according to the operation state, to obtain the overall energy efficiency of the wind turbines and the photovoltaic modules by cumulative summation, and to evaluate the energy efficiency of a single unit by time scale change.

[0063] In an optional embodiment, the power generation configuration module comprises:

[0064] a power generation prediction unit configured to acquire meteorological data in a preset historical time period of a wind energy storage power station and a light energy storage power station, to input the meteorological data into a wind resource power generation prediction model and a light resource power generation prediction model, and to obtain corresponding power generations;

[0065] The configuration model construction unit constructs an optimal configuration model with the maximum annual profit of the source network load storage life cycle as an objective function and power balance, new energy power station output and new energy power station local consumption as boundary conditions, to determine the power generation configuration of each new energy power station, considering the load electricity cost reduction and the corresponding on-grid income of the power generation of each new energy power station.

[0066] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, which are communicatively connected with each other, and the memory stores computer instructions; the processor executes the computer instructions to perform the multi-energy complementary new energy power station intelligent operation and maintenance method of the first aspect or any of the corresponding embodiments thereof.

[0067] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer instructions for making a computer execute the multi-energy complementary new energy power station intelligent operation and maintenance method of the first aspect or any of the corresponding embodiments thereof. BRIEF DESCRIPTION OF DRAWINGS

[0068] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0069] Figure 1 is a flowchart of the multi-energy complementary new energy power station intelligent operation and maintenance method of the embodiment of the present application;

[0070] Figure 2 is a structural block diagram of the multi-energy complementary new energy power station intelligent operation and maintenance system provided by the embodiment of the present application;

[0071] Figure 3 is a hardware structure schematic diagram of the computer device of the embodiment of the present application. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0073] According to the embodiment of the present application, a multi-energy complementary new energy power station intelligent operation and maintenance method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0074] In the present embodiment, a multi-energy complementary new energy power station intelligent operation and maintenance method is provided, which can be used in a computer device terminal, Figure 1 The flowchart of the multi-energy complementary new energy power station intelligent operation and maintenance method according to the embodiment of the present application is shown in Figure 1 The flowchart includes the following steps:

[0075] Step S101, based on the parameter data of a plurality of new energy power stations and the corresponding meteorological data, a power generation prediction model of each new energy power station is established to preset the power generation.

[0076] In the embodiment of the present application, the new energy power station includes a wind energy storage power station and a light energy storage power station, which is used as an example for illustration. Correspondingly, the power generation prediction model of the new energy power station includes a wind resource power generation prediction model and a light resource power generation prediction model, wherein: the wind resource power generation prediction model obtains the speed and turbulence intensity distribution in the wind farm based on the wind farm flow field numerical simulation model and the distribution of the unit, to predict the power generation of the wind turbine; the light resource power generation prediction model analyzes the influence of cloud clusters on solar radiation based on the digital twin model of the photovoltaic station equipment, and obtains the influence of cloud clusters on solar radiation distribution to predict the power generation of the photovoltaic component.

[0077] Further, the process based on the wind resource power generation prediction model in the embodiment of the present application includes:

[0078] A1, according to different arrangement modes l w , hub height h w , wind wheel diameter d w , the wind farm flow field numerical simulation model F is established:

[0079] F = ∑ (h w , d w , l w )

[0080] A2, based on the wind farm flow field numerical simulation model, the large eddy simulation method is used to obtain the speed and turbulence intensity distribution in the wind farm, and the average flow velocity of the wind turbine accessories is obtained according to the speed and turbulence intensity distribution in the wind farm;

[0081] The large eddy simulation method adopted in the embodiment of the present application is an important means for studying turbulent motion, which divides the turbulent motion into two parts of large scale and small scale, the small scale quantity is related to the large scale quantity through a model, and the large scale quantity is obtained through numerical calculation, and the large eddy simulation method is used to obtain the velocity and turbulence intensity distribution in the wind farm.

[0082] A3, according to the different arrangement modes of the wind turbine w , the average flow velocity of the wind turbine accessories v w and the wind energy conversion rate δ w The wind resource power generation prediction model is established to predict the power generation of the wind farm W w :

[0083] W w =∑(δ w ,v w ,l w ).

[0084] Further, the process of predicting the power generation of the photovoltaic module based on the light resource evaluation power prediction model includes:

[0085] B1, constructing a physical mechanism model of the photovoltaic module:

[0086]

[0087] Wherein, J (V) is the current density, J sc is the short circuit current, J0 is the reverse saturation current, A is the cross-sectional area of the solar cell, Rs is the series resistance, R SH is the parallel resistance, J is the current surface density, q is the charge unit, V is the voltage of the photovoltaic module, k B is the temperature coefficient of the solar panel, T a is the ambient temperature;

[0088] B2, measuring the cloud height and cloud thickness by using the sky imager, and fitting the relationship curve of the cloud thickness with time, cloud height and cloud movement trajectory;

[0089] B3, measuring the solar irradiance by using the irradiance table, and establishing the camera response function of the mapping relationship between the image gray value and the brightness value received by the camera according to the influence of the cloud, taking the natural logarithm of both sides to obtain:

[0090]

[0091] Wherein, Z is the pixel value of the picture, I is the scene brightness, Z and t are known, and I is unknown, under the assumption that the scene brightness is constant, E is proportional to t, for the i-th pixel, let I i =1, then t j =E iE is the brightness received by the camera, and a response curve is fitted;

[0092] B4, a photovoltaic module feature represented according to a physical mechanism model of the photovoltaic module and a conversion rate δ of irradiance s A light resource power generation prediction model is established to predict the power generation W of the photovoltaic power station s :

[0093] W s =∑(δ s ,J(V))。

[0094] In step S102, various types of unit operation related data of the new energy power station collected by the SCADA system and the CMS system are analyzed to obtain analysis data.

[0095] In actual application, the operation data of the power station is usually stored in a data acquisition and monitoring control system (SCADA system) and a condition monitoring system (CMS system). In the embodiment of the present application, the data stored in the two systems is cleaned, re-inducted and structured before data analysis. The specific analysis process is as follows:

[0096] C1, the SCADA system and the CMS system are divided into multiple time series according to a preset time resolution. In the embodiment of the present application, the time resolution is usually 15 minutes, and the length of the time series is selected as 1 hour according to experience.

[0097] C2, the similarity between the electrical data sequence of the unit measured and the electrical time sequence obtained by the upscaling method is calculated based on the Pearson correlation coefficient, and a similarity calculation result is obtained.

[0098] The Pearson correlation coefficient can define the similarity degree of two time series. The closer the value is to 1, the higher the similarity degree is. The definition is as follows:

[0099]

[0100] In the formula, r is the correlation coefficient of the two time series; is the average value of the two time series. Wherein X is the measured electrical data time sequence of the new energy power station, and Y is the electrical data time sequence of the new energy power station obtained by the upscaling method.

[0101] C3, the deviation degree between the electrical data sequence of the unit measured and the electrical time sequence obtained by the upscaling method is calculated based on the Euclidean distance, and a deviation degree calculation result is obtained. The Euclidean distance can intuitively measure the deviation degree of two time series. The greater the distance is, the higher the deviation degree is. The definition is as follows:

[0102] Let A, B be time series X=(x1, x2, …, xn ) on two fuzzy subsets, the Euclidean distance between A and B can be described by the following formula:

[0103]

[0104] The relative Euclidean distance between A and B is:

[0105]

[0106] When A and B take the same closed interval [alpha, beta], the Euclidean distance can be expressed as:

[0107]

[0108] It can be obtained:

[0109]

[0110] The Euclidean distance can be understood as the area between two time series, which can intuitively represent the deviation degree of two time series in the spatial scale. Wherein A is the reference measurement new energy power station electrical data time series, B is the new energy power station electrical data time series obtained by the upscaling method.

[0111] C4, based on the similarity calculation result and the deviation degree result, the key parameters of different unit operation states are extracted.

[0112] The embodiment of the application is based on the similarity calculation result and the deviation degree result, deeply mines the key parameters affecting the operation state of each unit of the new energy power station, studies the key influence parameter qualitative classification method based on the influence mode, and reveals the influence mechanism and coupling relationship of different levels of key influence parameters on the operation state of the wind turbine.

[0113] Step S103, based on the power generation prediction model and analysis data of each new energy power station, the fault model and the efficiency evaluation model of each type of unit are established, which are used for fault diagnosis and efficiency evaluation. The specific process is as follows:

[0114] D1, based on the key parameters of the operation state of the wind turbine and the photovoltaic component, the migration component analysis algorithm is used for distribution assimilation of different wind turbine and photovoltaic component operation data.

[0115] Because there are many distribution conditions of collected data in practice, such as time interval, sampling point, influence factor change rate, etc., the migration component analysis algorithm is used in this step to make the calculation value basically consistent with the distribution, which is called distribution assimilation. In the migration component analysis algorithm, it is assumed that the source domain D S ={X S ,Y S} is assumed, wherein X S is the source domain sample feature set, Y Sis a source domain sample label set; a target domain D T ={X T},wherein X T is a target domain sample feature set, and a target domain sample label set is unknown, wherein the source domain is a key influence parameter of a wind turbine or a photovoltaic module operating state, and the target domain is an operating state of the wind turbine or the photovoltaic module. S T S S T T

[0116] D2, based on a normal behavior model of the wind turbine or the photovoltaic module, taking the key influence parameter feature vector after distribution assimilation under all operating states as input, and obtaining model output power.

[0117] The normal behavior model of the wind turbine or the photovoltaic module in the embodiment of the application is a power output prediction model of the wind energy power station and the light energy power station under normal operation, which is obtained by training historical data, and is a relatively mature model, which will not be described here.

[0118] D3, using a random forest algorithm to evaluate indicators of model output power prediction residuals, and dividing different operating states of the wind turbine or the photovoltaic module. The random forest algorithm increases the difference between different classification models by randomly constructing different training sets, thereby improving the generalization ability and robustness of the combined classification model. Through k rounds of training, a multi-classification model system containing k models is obtained, and the final classification result of the system is determined by using a voting method. The final classification decision result, the application uses the random forest algorithm to divide the different operating states of the wind turbine or the photovoltaic module to obtain the fault and normal condition.

[0119] D4, obtaining the fault condition of a single unit according to the operating state, obtaining the overall energy efficiency of the wind turbine or the photovoltaic module by cumulative summation, and evaluating the energy efficiency of a single unit by time scale change.

[0120] Step S104, determining the power generation configuration of each new energy power station based on the predicted power generation of the power generation prediction model of each new energy power station. Specifically, it includes:

[0121] E1, obtaining the meteorological data of the wind energy power station and the light energy power station in a preset historical time period, inputting the meteorological data into the wind resource power generation prediction model and the light resource power generation prediction model to obtain the corresponding power generation. ​​​​​​

[0122] E2, considering the load electricity cost reduction and the corresponding on-grid income of the power generation of each new energy power station, constructing an optimization configuration model with the maximum of the whole life cycle annual income of the source network load storage as the objective function and the power balance, the new energy power station output and the local consumption of the new energy power station as the boundary conditions, and determining the power generation configuration of each new energy power station.

[0123] The objective function in the embodiment of the application is:

[0124] max F=C incom -C cost

[0125] F represents the annual net income, C incom represents the income generated each year, C cost represents the investment converted to each year.

[0126] The annual income of the source network load storage contains two parts: the reduction of the load electricity purchase cost and the new energy on-grid income:

[0127]

[0128] P load represents the load operation power; P buy represents the electricity purchase power; P sell represents the electricity sale power; γ t 1 large industrial electricity price; γ t 2 new energy on-grid electricity price; τ is a time interval, and T is 8760h.

[0129] The annual investment contains two parts, which are the equipment investment and the operation and maintenance cost, and the annual cost is converted as follows:

[0130] C cost =C eq +C op

[0131] C eq converted to the equipment investment each year, C op annual operation and maintenance cost.

[0132] The equipment annual investment is:

[0133]

[0134] c1 storage power and auxiliary equipment unit investment; c2 storage capacity unit investment; c3 new energy unit investment cost; c4 substation unit investment cost; P res rated power of storage; E res rated capacity of storage; P rtt rated power of substation; P rneNew energy installed power. r is the discount rate, n is the new energy life. Lambda is the battery equipment replacement cost coefficient. Among them, the annual operation and maintenance cost is:

[0135] C op = k1((c1P res + c2E res ) + c3P rne + c4P rtt )

[0136] K1 is the operation and maintenance coefficient.

[0137] The related formula of the boundary condition involved in the embodiment of the application is:

[0138] 1) Power balance

[0139] P ne (t) + P buy (t) + P dis (t) = P load (t) + P sell (t) + P ch (t)

[0140] P ne represents the new energy output power, P dis represents the energy storage discharge power, P ch represents the energy storage charging power, P load represents the load running power, P buy represents the power purchase, and P sell represents the power sale.

[0141] 2) New energy output

[0142] 0≤P ne (t)≤P rne

[0143] Among them, P rne represents the rated power.

[0144] 3) New energy local consumption rate

[0145] Q sell ≤(1-δ)Q ne

[0146]

[0147]

[0148] Delta is the new energy local consumption rate; Q ne is the annual power generation of new energy; Q sell is the annual power sale to the grid.

[0149] A multi-capacity complementary new energy power plant intelligent operation and maintenance system is also provided in the embodiment. The device is used to realize the above-mentioned embodiments and preferred embodiments, and has been described above. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware is also possible and contemplated.

[0150] The embodiment provides a multi-capacity complementary new energy power plant intelligent operation and maintenance system, as shown in the accompanying drawings, comprising: Figure 2

[0151] A power generation capacity prediction model establishing module 201 is configured to establish a power generation capacity prediction model for each new energy power plant based on parameter data of the new energy power plants and corresponding meteorological data, so as to preset the power generation capacity.

[0152] An operation data analysis module 202 is configured to analyze various types of operation-related data of each unit of the new energy power plants collected by the SCADA system and the CMS system, to obtain analysis data.

[0153] A fault model and performance evaluation model establishing module 203 is configured to establish a fault model and a performance evaluation model for each type of unit based on the power generation capacity prediction model and the analysis data of each new energy power plant, for fault diagnosis and performance evaluation.

[0154] A power generation configuration module 204 is configured to determine the power generation configuration of each new energy power plant based on the power generation capacity predicted by the power generation capacity prediction model of each new energy power plant.

[0155] In an optional embodiment, the power generation capacity prediction model establishing module comprises:

[0156] A wind resource power generation capacity prediction model constructing unit is configured to obtain the speed and turbulence intensity distribution in the wind farm based on a wind farm flow field numerical simulation model and the distribution of the units, to predict the power generation capacity of the wind turbines.

[0157] A light resource power generation capacity prediction model constructing unit is configured to analyze the influence of the cloud cluster on the solar radiation distribution based on a photovoltaic station equipment digital twin model, to predict the power generation capacity of the photovoltaic components.

[0158] In an optional embodiment, the operation data analysis module 202 comprises:

[0159] A time series dividing unit is configured to divide the operation-related data of each unit in the SCADA system and the CMS system into a plurality of time series according to a preset time resolution.

[0160] ​The similarity calculation unit is configured to calculate the similarity between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method based on the Pearson correlation coefficient, and obtain a similarity calculation result.

[0161] The deviation degree calculation unit is configured to calculate the deviation degree between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method based on the Euclidean distance, and obtain a deviation degree calculation result.

[0162] The key parameter extraction unit is configured to extract the key parameters of different operating states of the unit based on the similarity calculation result and the deviation degree result.

[0163] In an optional embodiment, the fault model and performance evaluation model establishment module 203 comprises:

[0164] The operating data distribution assimilation unit is configured to perform operating data distribution assimilation of different wind turbines and photovoltaic modules by using a transfer component analysis algorithm based on the key parameters of the operating states of the wind turbines and photovoltaic modules.

[0165] The output power acquisition unit is configured to acquire the model based on the normal behavior model of the wind turbines and photovoltaic modules, and taking the distribution-assimilated key influence parameter feature vectors in all operating states as input.

[0166] The operating state division unit is configured to divide the different operating states of the wind turbines and photovoltaic modules based on the model output power prediction residual error evaluation index by using a random forest algorithm.

[0167] The energy efficiency evaluation unit is configured to obtain the fault condition of a single unit according to the operating state, obtain the overall energy efficiency of the wind turbines and photovoltaic modules by cumulative summation, and evaluate the energy efficiency of a single unit by time scale change.

[0168] In an optional embodiment, the power generation configuration module 204 comprises:

[0169] The power generation prediction unit is configured to acquire the meteorological data in a preset historical time period of the wind energy storage power station and the light energy storage power station, and input the meteorological data into a wind resource power generation prediction model and a light resource power generation prediction model to obtain corresponding power generations.

[0170] The configuration model construction unit is configured to construct an optimization configuration model by taking the maximum of the life cycle annual revenue of the source, network, load and storage as an objective function, and taking power balance, new energy power station output and local consumption of the new energy power station as boundary conditions, and determine the power generation configuration of each new energy power station.

[0171] The further function descriptions of the above-mentioned modules and units are the same as those of the above-mentioned embodiments, and will not be described here again.

[0172] This invention also provides a computer device having the above-described features. Figure 2 The image shows a smart operation and maintenance system for a multi-energy complementary new energy power plant.

[0173] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 3 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 3 Take a processor 10 as an example.

[0174] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0175] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0176] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0177] The memory 20 can include a volatile memory, such as a random access memory, and / or can include a non-volatile memory, such as a flash memory, a hard disk or a solid state disk. The memory 20 can also include a combination of the above-mentioned types of memories.

[0178] The computer device also comprises a communication interface 30 for enabling the computer device to communicate with other devices or communication networks.

[0179] The embodiments of the present application also provide a computer readable storage medium, the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded from a network and stored in a local storage medium by original computer code, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor or programmable or special purpose hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that the computer, processor, microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, processor or hardware, the method shown in the above embodiments is implemented.

[0180] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A method for intelligent operation and maintenance of a multi-energy complementary new energy power plant, characterized in that, The method comprises the following steps: Based on the parameter data and corresponding meteorological data of multiple new energy power stations, a power generation prediction model of each new energy power station is established to preset the power generation, wherein the power generation prediction model of each new energy power station comprises a wind resource power generation prediction model and a light resource power generation prediction model, wherein: The wind resource power generation prediction model is based on a wind power base flow field numerical simulation model and the distribution of the units to obtain the speed and turbulence intensity distribution in the wind farm to predict the power generation of the wind turbine; the light resource power generation prediction model is based on a photovoltaic station equipment digital twin model to analyze the influence of cloud clusters on solar radiation to obtain the influence of cloud clusters on solar radiation distribution to predict the power generation of photovoltaic components; The method comprises the following steps: Based on the power generation prediction model of each new energy power station and the analysis data, a fault model and an efficiency evaluation model of each type of unit are established for fault diagnosis and efficiency evaluation, including: based on the key parameters of the running state of wind turbines and photovoltaic components, using the transfer component analysis algorithm to homogenize the running data distribution of different wind turbines and photovoltaic components; Based on the normal behavior model of wind turbines and photovoltaic components, the key influence parameter feature vector after distribution homogenization under all running states is input to obtain the model output power; Random forest algorithm is used to predict residual error evaluation index based on model output power to divide different running states of wind turbines and photovoltaic components; According to the running state, the fault condition of a single unit is obtained, the overall energy efficiency of wind turbines and photovoltaic components is obtained by cumulative summation, and the energy efficiency of a single unit is evaluated by time scale change; Based on the power generation prediction model of each new energy power station, the power generation configuration of each new energy power station is determined.

2. The method of claim 1, wherein, The multiple new energy power stations comprise wind energy storage power stations and light energy storage power stations.

3. The method of claim 1, wherein, The process based on the wind resource power generation prediction model comprises: According to different arrangement of wind farm l w , hub height h w , wind wheel diameter d w Establish wind power base flow field numerical simulation model F: Based on the wind power base flow field numerical simulation model, the large eddy simulation method is used to obtain the speed and turbulence intensity distribution in the wind farm, and the average flow velocity of the wind turbine accessories is obtained according to the speed and turbulence intensity distribution in the wind farm; According to the different arrangement of the wind turbines l w , the average flow velocity of the wind turbine accessories v w and the wind energy conversion rate A wind resource power generation prediction model is established to predict the power generation of the wind farm W w : 。 4. The method of claim 3, wherein, The process of predicting the power generation of photovoltaic components based on the light resource power generation prediction model comprises: A physical mechanism model of photovoltaic components is constructed: Where J(V) is the current density, J sc J0 is the short-circuit current, A is the cross-sectional area of ​​the solar cell, Rs is the series resistance, and R is the short-circuit current. SH Let J be the parallel resistance, J be the surface current density, q be the charge unit, V be the voltage of the photovoltaic module, and k be the voltage of the photovoltaic module. B T represents the temperature coefficient of the solar panel. a The ambient temperature; The cloud height and cloud thickness are measured by a sky imager, and the relationship curve of cloud thickness with time, cloud height and cloud motion trajectory is fitted; The solar irradiance is measured by an irradiance table, and the camera response function of the mapping relationship between the image gray value and the brightness value received by the camera is established according to the cloud influence, and the natural logarithm of both sides is taken to obtain Wherein, Z is the picture pixel value, I is the scene brightness, Z and t are known, I is unknown, under the assumption that the scene brightness is constant, E is proportional to t, for the ith pixel, let I i =1, then t j =E i , E is the brightness received by the camera, and the response curve is fitted. Photovoltaic module characteristics and irradiance conversion rates characterized according to a physical mechanism model of the photovoltaic module A light resource power generation prediction model is established to predict the power generation W of the photovoltaic power station s : 。 5. The method according to claim 1 or 2, characterized in that, The process of analyzing the running related data of multiple new energy power stations collected by the SCADA system and the CMS system to obtain analysis data comprises: The running related data of each unit in the SCADA system and the CMS system is divided into multiple time series according to the preset time resolution; The similarity between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method is calculated based on the Pearson correlation coefficient to obtain a similarity calculation result; The deviation degree between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method is calculated based on the Euclidean distance to obtain a deviation degree calculation result; The key parameters of different unit operating states are extracted based on the similarity calculation result and the deviation degree result.

6. The method of claim 1, wherein, The process of determining the power generation configuration of each new energy power station based on the predicted power generation of the power generation prediction model of each new energy power station, comprising: Obtain the corresponding power generation by inputting the meteorological data of the wind energy storage station and the light energy storage station in the preset historical time period into the wind resource power generation prediction model and the light resource power generation prediction model; Considering the load power consumption cost reduction and the corresponding on-grid income of the power generation of each new energy power station, taking the maximum annual life cycle income of source network load storage as the objective function, and taking power balance, new energy power station output and local consumption of new energy power station as boundary conditions to construct an optimization configuration model to determine the power generation configuration of each new energy power station.

7. A multi-energy complementary new energy power plant intelligent operation and maintenance system, characterized in that, The system comprises: A power generation prediction model establishment module is configured to establish a power generation prediction model of each new energy power station based on parameter data and corresponding meteorological data of a plurality of new energy power stations to predict power generation, The power generation prediction model establishment module comprises: A wind resource power generation prediction model construction unit is configured to obtain the speed and turbulence intensity distribution in the wind farm based on a wind farm flow field numerical simulation model and the distribution of units to predict the power generation of wind turbines; A light resource power generation prediction model construction unit is configured to analyze the influence of cloud clusters on solar radiation distribution based on a photovoltaic station equipment digital twin model to predict the power generation of photovoltaic components; An operation data analysis module is configured to analyze various types of unit operation related data collected by the SCADA system and the CMS system of a plurality of new energy power stations to obtain analysis data; A fault model and performance evaluation model establishment module is configured to establish a fault model and a performance evaluation model for each type of unit based on the power generation prediction model of each new energy power station and the analysis data, which are used for fault diagnosis and performance evaluation, comprising: An operation data distribution assimilation unit is configured to use a migration component analysis algorithm to assimilate the operation data distribution of different wind turbines and photovoltaic components based on the key parameters of the operating states of wind turbines and photovoltaic components; An output power acquisition unit is configured to acquire a model based on the normal behavior model of wind turbines and photovoltaic components, with the feature vector of the key influence parameters after distribution assimilation in all operating states as input; An operating state division unit is configured to divide the operating states of wind turbines and photovoltaic components based on the model output power prediction residual error evaluation index using a random forest algorithm; An energy efficiency evaluation unit is configured to obtain the fault condition of a single unit according to the operating state, to obtain the overall energy efficiency of wind turbines and photovoltaic components by cumulative summation, and to evaluate the energy efficiency of a single unit by time scale change. The power generation configuration module is configured to determine the power generation configuration of each new energy power station based on the predicted power generation of each new energy power station by the power generation prediction model.

8. The system of claim 7, wherein, The operation data analysis module includes: The time series division unit is configured to divide each unit operation related data in the SCADA system and the CMS system into a plurality of time series according to a preset time resolution. The similarity calculation unit is configured to calculate the similarity between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method based on the Pearson correlation coefficient, to obtain a similarity calculation result. The bias degree calculation unit is configured to calculate the bias degree between the measured electrical data sequence of the unit and the electrical time sequence obtained by the upscaling method based on the Euclidean distance, to obtain a bias degree calculation result. The key parameter extraction unit is configured to extract the key parameters of different unit operation states based on the similarity calculation result and the bias degree result.

9. The system of claim 8, wherein, The power generation configuration module includes: The power generation prediction unit is configured to obtain the corresponding power generation by inputting the meteorological data of the wind energy storage power station and the light energy storage power station in a preset historical time period into the wind resource power generation prediction model and the light resource power generation prediction model. The configuration model construction unit is configured to consider the load power consumption cost reduction and the corresponding grid access income of the power generation of each new energy power station, to construct an optimal configuration model with the maximum source network load storage life cycle annual income as an objective function and the power balance, the new energy power station output and the local consumption of the new energy power station as boundary conditions, to determine the power generation configuration of each new energy power station.

10. A computer device, comprising: It includes: The memory and the processor are in communication connection with each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the multi-energy complementary new energy power station intelligent operation and maintenance method of any one of claims 1-6.

11. A computer readable storage medium, characterized in that, The computer readable storage medium stores computer instructions, and the computer instructions are used to make the computer execute the multi-energy complementary new energy power station intelligent operation and maintenance method of any one of claims 1-6.

Images