Summary of the invention
Goal of the invention: the object of the invention is for the deficiencies in the prior art, relates to the dynamic frequency-conversion collecting method of a kind of wind-solar-storage joint generating state monitoring.
Therefore need to set up one for above-mentioned situation and meet wind-solar-storage joint generating set syntype, can according to real system ruuning situation self-adaptative adjustment data acquisition session state, while meeting data integrity and real-time, make full use of the collecting method of Internet resources.
Technical scheme: the dynamic frequency-conversion collecting method of a kind of wind-solar-storage joint generating state monitoring of the present invention, combined generating system comprises mechanical floor, acquisition layer, network layer and supervisory layers, described mechanical floor comprises wind energy turbine set, photovoltaic DC field and energy storage device, described acquisition layer comprises the level work station, wind energy turbine set field communicated to connect with described wind energy turbine set, the level work station, field, photovoltaic field communicated to connect with described photovoltaic DC field, the level work station, energy storage device field communicated to connect with described energy storage device, described supervisory layers comprises dispatching and monitoring work station between field, between described field, dispatching and monitoring work station is communicated to connect by described network layer and described acquisition layer, carry out as follows:
Step1: between the field of supervisory layers, dispatching and monitoring work station obtains combined generating system and runs integrated mode, simultaneously initialization task characteristic parameter, initialized task characteristic parameter comprises each rank frequency acquisition and the at different levels total meritorious minute levels corresponding with the granularity that described each rank frequency acquisition divides and to fluctuate threshold value;
Step2: whether supervisory layers monitoring reaches the image data time, when arriving the image data time, send acquisition time instruction to acquisition layer, the level work station, each field of acquisition layer gathers the active power p communicating with each on-site individuality of connection, dispatching and monitoring work station between the field active power collected being sent to supervisory layers;
Step3: between field, dispatching and monitoring work station is to the active power p process received, and extracts maximum active-power P in t minute
max, twith t minute in minimum active-power P
min, t, and calculate average active power P in t minute
avg, t, then calculate total meritorious minute level fluctuation ratio a
1, computational methods are
Step4: between field, dispatching and monitoring work station is analyzed each acquisition layer according to the active power p received, for being in the acquisition layer producing active power state, will always gain merit a minute level fluctuation ratio a
1compare with total meritorious minute level threshold value that fluctuate, adopt double threshold multilevel threshold frequency modulation mechanism to judge whether to need to change frequency acquisition, if need to change frequency, determine the frequency acquisition after change and be sent to level work station, each field.
Being further defined to of technical solution of the present invention, the operation integrated mode described in Step1 comprises that wind energy turbine set is exerted oneself separately, exert oneself separately in photovoltaic field, wind-powered electricity generation/photovoltaic field combine exert oneself, wind-powered electricity generation/energy storage field combine exert oneself, photovoltaic/energy storage field combines and to exert oneself or one or more the combination in exerting oneself is combined in wind-powered electricity generation/photovoltaic/energy storage field.
Further, the concrete grammar of the employing double threshold multilevel threshold frequency modulation mechanism described in step Step4 is: first, during initialization initialization task characteristic parameter, initialized each rank frequency acquisition is F
i, corresponding with each rank frequency acquisition total meritorious minute level threshold value that fluctuate is that 2n is individual, and wherein n Low threshold is expressed as α
iL(1≤i≤n), n high threshold is expressed as α
iH(1≤i≤n), and meet α
iL< α
iH< α
(i+1) L(1≤i≤n); Then, as total meritorious minute level fluctuation ratio α
1< α
iLtime, between field dispatching and monitoring work station by frequency acquisition by F
ibe down to F
i-1; As total meritorious minute level fluctuation ratio a
1> a
iHtime, between field dispatching and monitoring work station by frequency acquisition by F
irise to F
i+1.
Further, in step Step4, for being in the acquisition layer not producing active power state, under original frequency acquisition, decline frequency acquisition is to the minimum frequency acquisition of initial setting up step by step.
Beneficial effect: the dynamic frequency-conversion collecting method of a kind of wind-solar-storage joint generating state monitoring provided by the invention, in conjunction with wind-solar-storage joint electricity generation system combined running pattern and service data variation characteristic, adopt layered distribution type scheduling strategy, realize the execution frequency of self adaptation dynamic conditioning acquisition tasks.Empirical tests, under identical frequency acquisition, according to the composes curve that self-adapting frequency conversion algorithm collects than fixed frequency method, precision improves an order of magnitude, effectively reduces the distortion factor of signal data acquisition; Under identical acquisition precision, the frequency acquisition of self-adapting frequency conversion acquisition method is lower than fixed frequency acquisition method by more than 30%, and it is 1/3rd that the data volume of its transmission decreases, and significantly improves network transmission bandwidth utilance; The present invention builds data acquisition session parallel compensate model, application simulation annealing algorithm obtains optimal solution and the suboptimal solution of system acquisition frequency set in finite time, realize the optimization that mission frequency distributes, obtain and make the dynamic frequency allocation scheme that task execution time is minimum, network energy consumption is minimum; The present invention adopts double threshold multilevel threshold mechanism to realize frequency acquisition handoff procedure, solve jitter problem when adopting single threshold frequency to switch, also be better than traditional double threshold method simultaneously, reduce the difference between side frequency, identifiable design goes out data variation more by a small margin, achieve between frequency and steadily switch, greatly reduce the impact of threshold value value on energy-saving effect.
Embodiment 1: a kind of dynamic frequency-conversion collecting method of wind-solar-storage joint generating state monitoring, the structural representation of combined generating system as shown in Figure 1, system adopts cross direction profiles, longitudinal layered thought, set up distributed network hardware platform, comprise mechanical floor, acquisition layer, network layer and supervisory layers, described mechanical floor comprises wind energy turbine set, photovoltaic DC field and energy storage device, described acquisition layer comprises the level work station, wind energy turbine set field communicated to connect with described wind energy turbine set, the level work station, field, photovoltaic field communicated to connect with described photovoltaic DC field, the level work station, energy storage device field communicated to connect with described energy storage device, described supervisory layers comprises dispatching and monitoring work station between field, between described field, dispatching and monitoring work station is communicated to connect by described network layer and described acquisition layer.
Described mechanical floor, according to generation mode, is divided into multiple region to carry out distributed capture whole power plant zoning, comprises wind energy turbine set, photovoltaic cell, and energy-storage system
Described acquisition layer, comprises level work station, field set in each region, carries out dispatching and monitoring to the acquisition tasks of generating field in region.
Described networking layer, adopts Ethernet mode to set up data acquisition system, and communication connects employing ICP/IP protocol.The unified telemechanical agreement based on TCP is followed in remote action data transmission in supervisory layers, ensures the compatibility of system communication, reliability and opening.
Described supervisory layers, is responsible for the real-time monitoring and scheduling of system running state.Comprise dispatching and monitoring work station between field, failure warning management work station, system maintenance work station and database server.
Wind-solar-storage joint electricity generation system combined running pattern and service data variation characteristic, the present invention arranges Control and Schedule work station between field in supervisory layers and performs self adaptation dynamic frequency-conversion data collection strategy, carries out dynamic adjustments to the acquisition tasks of each place.
The flow chart of the dynamic frequency-conversion collecting method of wind-solar-storage joint generating state monitoring as shown in Figure 2, carries out as follows:
Step1: between the field of supervisory layers, dispatching and monitoring work station obtains combined generating system and runs integrated mode, simultaneously initialization task characteristic parameter, initialized task characteristic parameter comprises each rank frequency acquisition and the at different levels total meritorious minute levels corresponding with the granularity that described each rank frequency acquisition divides and to fluctuate threshold value.
Described operation integrated mode comprises that wind energy turbine set is exerted oneself separately, exert oneself separately in photovoltaic field, wind-powered electricity generation/photovoltaic field combine exert oneself, wind-powered electricity generation/energy storage field combine exert oneself, photovoltaic/energy storage field combines and to exert oneself or one or more the combination in exerting oneself is combined in wind-powered electricity generation/photovoltaic/energy storage field.
Combining wind and light to generate electricity has natural complementarity to a certain degree, by its smoothing adjustment of energy-storage system.For one day 24 hours, retrain by environmental condition, wind power generation amount is the feature that daytime, night was many less, and only has when wind speed is between minimum threshold wind velocity, excision wind speed, and wind power system just has output; And photovoltaic generation is subject to the impact of the factor such as solar radiation, ambient temperature, only has by day during solar irradiation and just can exert oneself.Consider the factors of limit life of energy-storage battery, scene exert oneself cannot reach grid-connected requirement time energy-storage system regulate, otherwise energy-storage system is failure to actuate.Wind-solar-storage joint power-generating control system, according to operation plan, scene prediction, carries out overall view monitoring to combined generating system, achieves 6 kinds of different configuration operational mode seamless switchings.As shown in table 1.P
pv, P
wd, P
batphotovoltaic module power output valve, Wind turbines power stage value, energy storage device power stage value respectively.
Table 1 wind-light storage combined running modes relationships figure
Operational mode |
Outside environmental elements |
Energy storage regulates |
Gross capability P |
Wind is exerted oneself separately |
Wind speed is can meet grid-connected conditions, unglazed photograph in range of operation |
No |
P
wd |
Wind/storage is combined and is exerted oneself |
Wind speed is not can meet grid-connected conditions, unglazed photograph in range of operation |
Be |
P
wd+P
bat |
Light is exerted oneself separately |
Have illumination and meet grid-connected conditions, wind speed is can outside range of operation |
No |
P
pv |
Light/storage is combined and is exerted oneself |
Have illumination and do not meet grid-connected conditions, wind speed is can outside range of operation |
Be |
P
pv+P
bat |
Wind/light is combined and is exerted oneself |
Have illumination, wind speed can in range of operation, synthesis is exerted oneself and is met grid-connected conditions |
No |
P
pv+P
wd |
Wind-solar-storage joint is exerted oneself |
Have illumination, wind speed can in range of operation, synthesis is exerted oneself discontented sufficient grid-connected conditions |
Be |
P
pv+P
wd+P
bat |
Step2: whether supervisory layers monitoring reaches the image data time, when arriving the image data time, send acquisition time instruction to acquisition layer, the level work station, each field of acquisition layer gathers the active power p communicating with each on-site individuality of connection, dispatching and monitoring work station between the field active power collected being sent to supervisory layers.
For each generating place, the present invention arranges a level work station respectively at wind energy turbine set, photovoltaic cell level and energy-storage system, to image data classification in region, and determines to gather by demand the task start mode that all types of data adopt.Define the characteristic parameter (as time of implementation, time limit, cycle etc.) of task during beginning, and determine each task priority; Then, the task of selecting dynamic priority the highest is its Resources allocation, and responds task requests, and data acquisition thread starts to perform this task.Timing carries out dynamic priority renewal to the task in task queue, upgrades in the task scheduling point timing of setting the dynamic priority calculating all ready tasks, sorts and select to perform head of the queue task to task.When having new task to arrive, calculate its priority, and insert the relevant position in task queue.When the priority value of multiple task is equal, then according to certain priority with reference to its characteristic parameter, the priority as characteristic parameter is from high to low: time limit, time of implementation, cycle.
Step3: for the system being in generating state, the situation of change of fluctuation ratio α that always gains merit in the recent period according to it, judges whether to change frequency.If α exceedes or close on threshold value lower than what set, then generate a frequency acquisition altering event and frequency acquisition is adjusted accordingly, otherwise maintain original frequency acquisition.
Between, dispatching and monitoring work station is to the active power p process received, and extracts maximum active-power P in t minute
max, twith t minute in minimum active-power P
min, t, and calculate average active power P in t minute
avg, t, then calculate total meritorious minute level fluctuation ratio a
1, computational methods are
For the system being in the state of not exerting oneself, under original acquisition state, be down to minimum frequency acquisition F1 step by step.Definition frequency acquisition can only switch between adjacent frequency, switching of can not bypassing the immediate leadership, and F1≤Fi≤Fn+1.Wherein, F1 is the minimum frequency acquisition of default, and Fn+1 is the maximum frequency acquisition of default.
Step4: between field, dispatching and monitoring work station is analyzed each acquisition layer according to the active power p received, for being in the acquisition layer producing active power state, will always gain merit a minute level fluctuation ratio a
1compare with total meritorious minute level threshold value that fluctuate, adopt double threshold multilevel threshold frequency modulation mechanism to judge whether to need to change frequency acquisition, if need to change frequency, determine the frequency acquisition after change and be sent to level work station, each field.
Described employing double threshold multilevel threshold frequency modulation mechanism, as shown in Figure 3, concrete grammar is its structural representation: first, and during initialization initialization task characteristic parameter, initialized each rank frequency acquisition is F
i, corresponding with each rank frequency acquisition total meritorious minute level threshold value that fluctuate is that 2n is individual, and wherein n Low threshold is expressed as α
iL(1≤i≤n), n high threshold is expressed as α
iH(1≤i≤n), and meet α
iL< α
iH <α
(i+1) L(1≤i≤n); Then, as total meritorious minute level fluctuation ratio α
1< α
iLtime, between field dispatching and monitoring work station by frequency acquisition by F
ibe down to F
i-1; As total meritorious minute level fluctuation ratio α
1> α
iHtime, between field dispatching and monitoring work station by frequency acquisition by F
irise to F
i+1.For being in the acquisition layer not producing active power state, under original frequency acquisition, decline frequency acquisition is to the minimum frequency acquisition of initial setting up step by step.
Acquisition tasks is performed efficiently and in real time for realizing each task of data acquisition system, effectively promote network performance simultaneously, the present invention adopts simulated annealing, using time overhead and network overhead as evaluation model, in conjunction with wind-solar-storage joint electricity generation system combined running pattern and service data variation characteristic, formulating target function E (x) of Model for Multi-Objective Optimization, is dynamically that frequency acquisition is distributed in collection point, each region.
One, dynamic frequency-conversion data acquisition session parallel compensate model
1, mission frequency apportion model evaluation function
(1) time overhead, comprising: interaction time between field level acquisition time, field
Time is the time overhead of layered distribution type data acquisition; N is place quantity,
it is the acquisition node number of i-th place; Time
ijfor gathering each node required time, time
interfor the mutual required time of control centre of work station between this place and field.
(2) network overhead, comprising: communication energy consumption, acquisition node process energy consumption between field level communication energy consumption, field
C is the network overhead of layered distribution type data acquisition; N is place quantity,
with
be intra-area communication expense and the inter-domain communication expense of i-th place,
for the energy ezpenditure of node processing task.
2, Model for Multi-Objective Optimization
For obtaining the dynamic frequency allocation scheme meeting data acquisition session real-time and network bandwidth requirement, the present invention, in conjunction with above-mentioned evaluation function, to every optimizing index weighted comprehensive, obtains the target function of following Model for Multi-Objective Optimization:
E(x)=w
1×min(time)+w
2×min(4)
W wherein
ifor can value be the weight function of zero.
Two, algorithm is arranged
If the frequency acquisition that data acquisition equipment can provide integrates as D={d1, d2 ..., dm}.Fixed Frequency Assignment is exactly complete sharing out the work to each device frequency in a fixing frequency set, makes that task execution time is minimum, network energy consumption is minimum.X represents frequency allocation plan, be expressed as X=(x1, x2 ..., xN), collector quantity in N expression system, xi arranges with priority rule.
New explanation production method: the present invention adopts the new explanation production method in two migration field, produces new parallel compensate solution by changing two parameters at every turn: Xold and Xnew represents the parallel compensate solution before and after conversion respectively.
State accepts function: effectively avoid being absorbed in local suboptimal solution.The present invention adopts min{1, exp (-△ E/t) } > random (0,1) is as the condition accepting new state.
Controling parameters t: the process controlling whole annealing algorithm, determines the iterations separated at each " temperature ".The present invention adopts geometry Annealing Strategy, and namely the decline of temperature linearly declines, and the iterative steps in each temperature is identical.Temperature declines and meets following relation:
t
k+1=α×t
k(0≤α≤1)(5)
Annealing scheme: set total temperature decline number of times as definite value K, when temperature iterations reaches K, stops computing.Further, the iterative steps of each temperature is by accepting rate control.Given one accepts ratio indicator R (0.9), iteration step length upper limit U (100) and lower limit L (10), each temperature at least iteration L time, and record the total degree of same temperature iteration and received number of times, when iterations is more than L, if accept number of times when being not less than R with the ratio of total degree, in this temperature no longer iteration and start temperature and reduce, otherwise iterate to upper limit step number always.
Three, algorithm steps
With the place that generates electricity for allocation unit, list all acquisition tasks according to priority orders.
According to priority order be from high to low to acquisition tasks distribute perform frequency, until all mission frequencys are assigned.
The task solution being positioned at high priority fixed, other separate the initial solution of element as simulated annealing.
Perform simulated annealing, find the solution meeting constraints, merge with the solution that high-priority task is distributed before, the set of composition last solution, obtains final scheme.
Above-mentioned simulated annealing concrete steps:
Select the solution element i0 except high priority; I:=i0; K:=0; T0:=tmax (initial temperature).
If reach Inner eycle stop condition changing temperature, then arrive step3;
Otherwise, produce Xnew, calculating target function Enew, by △ E=Enew-Eold by current solution Xold through conversion; If △ E≤0, then Eold:=Enew, Xold:=Xnew, if otherwise exp (-△ E/tk) > random (0,1) time, then Eold:=Enew, Xold:=Xnew; Repeat step 2.
Tk+1:=d (tk); K:=k+1; If meet stop condition, stop calculating; Otherwise, get back to step 2.
By above-mentioned algorithm, the frequency allocation plan of level work station, each field acquisition tasks in units of the place that generates electricity can be obtained.Such as there is allocative decision (2,3,1), then represent that the frequency acquisition that level work station, field is distributed is d2, d3, d1 respectively.
As shown in Figure 4, by emulation experiment, simulate the effect contrast figure of self adaptation dynamic frequency-conversion collecting method and fixing frequency acquisition collecting method.
As shown in Figure 5, by emulation experiment, for one day 24 hours, simulate self adaptation dynamic frequency-conversion collecting method be applied to wind-solar-storage joint generating state monitoring in simulated effect figure.
Above-mentioned simulation result shows, a kind of self adaptation dynamic frequency-conversion collecting method being applied to the monitoring of wind-solar-storage joint generating state of the present invention, in time can regulate frequency acquisition, compared with fixing frequency acquisition, under the requirement meeting task execution time and network energy consumption, effectively reduce the distortion factor of sampling, improve the stability of a system.
As mentioned above, although represented with reference to specific preferred embodiment and described the present invention, it shall not be construed as the restriction to the present invention self.Under the spirit and scope of the present invention prerequisite not departing from claims definition, various change can be made in the form and details to it.