WO2020119012A1 - 基于历史数据斜坡响应的动态系统静态增益估计方法 - Google Patents
基于历史数据斜坡响应的动态系统静态增益估计方法 Download PDFInfo
- Publication number
- WO2020119012A1 WO2020119012A1 PCT/CN2019/085188 CN2019085188W WO2020119012A1 WO 2020119012 A1 WO2020119012 A1 WO 2020119012A1 CN 2019085188 W CN2019085188 W CN 2019085188W WO 2020119012 A1 WO2020119012 A1 WO 2020119012A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- output
- input
- data
- data segment
- static gain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Definitions
- the present disclosure relates to the technical field of industrial big data analysis, and particularly to a static system static gain estimation method based on historical data slope response.
- Static gain is important information for dynamic systems. Static gain of dynamic systems is usually used to design feedback controllers, monitor process changes, and optimize operating performance. Static gain can be obtained in the design stage of dynamic systems, and the design stage and actual operating conditions are different. Therefore, the static gain of dynamic systems is often unavailable in practice, and must be estimated from observed data samples. A common method is to estimate the static gain of the dynamic system from the steady state values of the input and output or some special types of tests.
- the existing method has two limitations. First, data samples under steady-state conditions are difficult to obtain. For some dynamic systems, special types of tests are not allowed. Therefore, it is advisable to estimate the static gain from data samples collected from daily operations. Second, the system identification technique is based on a hypothetical condition. The hypothetical model set is rich enough to contain the real model. Moreover, this hypothesis cannot be verified in practice, nor can it find the deviation between the estimated static gain and the actual gain.
- the present disclosure provides a dynamic system static gain estimation method based on the slope response of historical data to estimate the static gain of the dynamic system from relevant information hidden in industrial big data. This method does not rely on data analysis, and also avoids the situation where the deviation between the estimated static gain and the actual gain cannot be determined. It has good application value in model estimation and can overcome the uncertainty of system identification.
- the present disclosure provides a dynamic system static gain estimation method based on historical data slope response
- Dynamic system static gain estimation method based on historical data slope response including:
- a piecewise linear representation method is used to divide the output time series of the dynamic system into several output time segments; meanwhile, the input time series of the dynamic system is divided into several input data segments; each data segment is represented by a straight line;
- the static gain is estimated through the collection of data segments with significant amplitude changes.
- the specific steps of dividing the output time series of the dynamic system into several output time periods are:
- n m represents the first data sample in the mth segment, m ⁇ [1, M], and n m+1 -n m -1 is the total number of samples in the mth data segment;
- a m represents the initial value of the m-th data segment
- b m represents the slope of the m-th data segment
- e(n) is interference
- L(M) is the fitting error loss function
- the input time series of the dynamic system is divided into several input data segments:
- the piecewise linear representation method is used to input the time series Split into M data segments sequentially The number of segments M, the estimated value of M:
- a y,0 is the threshold of significant amplitude change of y
- the specific steps for estimating the static gain are:
- Step 3.1 Estimate by least square method The static gain K 1 , K 2 , ... K I ;
- K is an I-dimensional vector composed of K 1 , K 2 , ... K I , and the estimate of K It is obtained by the least square method:
- Step 3.2 Find out A y,l and its estimated value The maximum deviation of the data segment in paragraph l 0, if the amplitude variation period satisfies the inequality:
- ⁇ y is a parameter selected by the user, indicating with The acceptable level of maximum deviation.
- Step 3.3 Repeat steps 3.1 and 3.2 until no data segment with amplitude deviation greater than ⁇ y is found; at this time, the estimated static gain vector is expressed as Where S 1 is the set with a significant change in amplitude at the end of the current step, ie
- Step 3.4 The remaining set of data segments with significantly varying amplitudes are:
- the beneficial effect of the present disclosure is to verify the effectiveness of the method through visualization, and overcome the problem that it is difficult to verify the static gain estimation using the system identification method.
- FIG. 1 is a flow chart of static gain estimation of a dynamic system based on slope response in industrial large numbers according to the present invention
- FIG. 2(a)-FIG. 2(d) are sample diagrams of sampled data in a specific implementation example of the present invention.
- 3(a)-FIG. 3(f) are graphs of calculation data in a specific implementation example of the present invention.
- the dynamic system refers to a system whose state changes with time.
- the static gain refers to the degree of unit change of the system from one steady state to a new steady state.
- the ramp response refers to the time response caused by the change slope of an input quantity from zero to a certain finite value.
- the output time series refers to a sequence of output values arranged in chronological order.
- the input time sequence refers to a sequence of input variable values arranged in chronological order.
- the static system static gain estimation method based on historical data slope response includes:
- Step 1 Divide the time series of output y and input u i into short data segments, and each data segment is represented by a straight line.
- n m represents the first data sample in the mth segment, m ⁇ [1, M], and n m+1 -n m -1 is the total number of samples in the mth data segment;
- a m represents the initial value of the m-th data segment
- b m represents the slope of the m-th data segment
- e(n) is interference
- L(M) is the fitting error loss function
- Step 2 From the data segment, find the slope response where the input and output are both on a straight line and the amplitude changes greatly.
- a y,0 is the threshold of y significant amplitude change.
- D m Construct a determination coefficient D m , the closer the value of D m is to 1, the higher the fit.
- Step 3 Estimate the static gain from the slope response with significant amplitude changes at the input and output.
- Step 3.1 By solving multiple linear equations, estimate The static gains K 1 , K 2 ,...K I.
- K is an I-dimensional vector composed of K 1 , K 2 , ... K I , and the estimate of K is obtained by the least square method:
- the confidence interval of K can be estimated.
- Step 3.2 Find the amplitude change A y,l and its estimated value The data segment l 0 with the largest deviation between, if the amplitude variation of this segment satisfies the inequality:
- ⁇ y is a parameter selected by the user, indicating with The acceptable level of the maximum difference between.
- Step 3.3 Repeat steps 3.1 and 3.2 until no data segment with amplitude variation deviation greater than ⁇ y is found.
- the estimated static gain vector is expressed as Where S 1 is the set with a significant change in amplitude at the end of this step, ie
- Step 3.4 At this time, the remaining data sets with significantly varying amplitudes are:
- results of each step in this embodiment can be verified by visual inspection of related graphics.
- the found slope response can be verified by checking whether the input and output data segments are on a straight line.
- the purpose of this embodiment is to provide a method for estimating the static gain of a large 300MW coal-fired generating unit.
- this embodiment provides a method for estimating the static gain of a large-scale 300MW coal-fired generating unit, including the following steps:
- the output and input time series are divided into short data segments, and each data segment is represented by a straight line;
- the static gain of coal-fired generating units is estimated from the slope response with significant amplitude changes at the input and output.
- the active power (y) data samples in the unit were collected as the system output, the main steam flow controller output (u 1 ) and the main The steam pressure (u 2 ) data sample is used as the system input.
- U estimated static gain G 1 and y is 1 and the static gain G u 2 and y 2.
- the first step is to segment a one-hour data sample using a piecewise linear representation method.
- Figure 2(a) and Figure 2(b) are The timing diagram of Indicates that the one-hour data sample is divided into 3 segments. Similarly, you can get with The segmentation results are shown in Figure 2(c) and Figure 2(d).
- the thresholds of u 1 and u 2 can be calculated as with
- Table 1 shows the amplitude variation of the data segment of output y and input u 1 and u 2 .
- the changes in amplitude are greater than A y,0 ; for u 1 in Figure 2(c), the changes in the first three amplitudes are greater than
- According to the total instruction sequence I(n), three data segments with significant amplitude changes at both input and output are obtained: [1,658], [1136,1677] and [1678,2305]. get
- the third step from the collection Estimate the static gain group in Figure 3(a), Figure 3(b) and Figure 3(c), select A y, l and Acceptable level of maximum deviation between Table 2 gives three sets of static gain estimates and their confidence intervals. As shown in Figure 3(d), Figure 3(e) and Figure 3(f), A y, l and The deviation between them is less than ⁇ y .
- the purpose of this embodiment is to provide a feedback controller design method based on the static gain estimation method described in Embodiment 2. It includes the following steps:
- the output and input time series are divided into short data segments, and each data segment is represented by a straight line;
- the static gain is used in the feedback controller design of the generator set.
- the feedback controller C(s) is limited to PI form, then the feedback controller can be expressed as Assuming a first-order delayed system
- the time constant T and the delay ⁇ are known.
- the feedback controller design method includes the following steps:
- the time series of output y and input u is divided into short data segments, and each data segment is represented by a straight line;
- the static gain K of the dynamic system is estimated from the slope response of the input and output with significant amplitude changes; from the known system time constant T and delay ⁇ , and the obtained static gain K; using the Ziegler-Nichols parameter tuning method, based on The empirical tuning formula is used to calculate the feedback controller parameters K p and T i ;
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Complex Calculations (AREA)
- Feedback Control In General (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本发明涉及基于历史数据斜坡响应的动态系统静态增益估计方法。该方法首先要完成对历史数据的分析,采用分段线性表示方法将输入和输出的时间序列分割成短数据段。其次,找出输入和输出同时呈直线且处于振幅变化较大的斜坡响应,较大幅度变化的阈值是根据振幅变化与确定系数之间的关系确定的。最后通过求解斜坡响应幅值变化较大的多个线性方程,得到估计的静态增益。本发明可以通过可视化验证该方法的有效性,克服了使用系统识别方法难以验证静态增益估计的问题。从而避免了估计的静态增益和实际增益之间偏差无法判断的情况。
Description
本公开涉及工业大数据分析技术领域,特别是涉及基于历史数据斜坡响应的动态系统静态增益估计方法。
本部分的陈述仅仅是提高了与本公开相关的背景技术,并不必然构成现有技术。
静态增益是动态系统的重要信息,动态系统的静态增益通常用于设计反馈控制器、监测过程变化和优化运行性能,静态增益可在动态系统设计阶段获得,而设计阶段和实际操作的条件不同,因此,动态系统的静态增益在实践中往往是不可用的,必须从观察到的数据样本中估计。一种常用的方法就是从输入和输出的稳态值或某些特殊类型的测试中估计动态系统的静态增益。
现有方法存在两个局限性。首先,稳态条件下的数据样本很难获得,对于某些动态系统,不允许做特殊类型的测试。因此,从日常操作收集的数据样本来估计静态增益是可取的。第二,系统辨识技术是基于一个假设条件,假设模型集足够丰富,足以包含真实模型;而且,这种假设在实践中无法证实,也无法找到估计的静态增益和实际增益之间的偏差。
发明内容
为了解决现有技术的不足,本公开提供了基于历史数据斜坡响应的动态系统静态增益估计方法,从隐藏在工业大数据中的相关信息估计动态系统的静态增益。此方法不依赖数据分析,而且也避免了估计的静态增益和实际增益之间偏差无法判断的情况。在模型估计方面有很好的应用价值,可以克服系统辨识的不确定性。
第一方面,本公开提供了基于历史数据斜坡响应的动态系统静态增益估计方法;
基于历史数据斜坡响应的动态系统静态增益估计方法,包括:
采用分段线性表示方法将动态系统的输出时间序列分割成若干个输出时间段;同时,将动态系统的输入时间序列分割成若干个输入数据段;每个数据段均用一条直线表示;
找出输入值和输出值同时均处于各自的直线数据段上且输入值和输出值的振幅变化均超过设定阈值的数据段,此数据段即为目标斜坡响应,将找出的所有输入数据段和所有输出数据段组成具有显著振幅变化的数据段集合;
通过具有显著振幅变化的数据段集合,估计静态增益。
作为一种可能的实现方式,所述将动态系统的输出时间序列分割成若干个输出时间段的 具体步骤为:
用线性回归模型来描述:
y(n)=a
m+b
mn+e(n);
其中,a
m表示第m段数据段的初始值,b
m表示第m段数据段的斜率,e(n)是干扰;
其中,L(M)为拟合误差损失函数:
作为一种可能的实现方式,将动态系统的输入时间序列分割成若干个输入数据段:
作为一种可能的实现方式,找出输入值和输出值同时均处于各自的直线数据段上且输入值和输出值的振幅变化均超过设定阈值的数据段,此数据段即为目标斜坡响应,将找出的所有输入数据段和所有输出数据段组成具有显著振幅变化的数据段集合的具体步骤为:
对于输出来说,计算第m段的振幅变化量:
引入序列:
其中,A
y,0为y显著幅度变化的阈值;
整体序列:
作为一种可能的实现方式,通过具有显著振幅变化的数据段集合,估计静态增益的具体步骤为:
从K的高斯分布,估计K的置信区间;
步骤3.4:剩余的具有显著变化幅值的数据段集合为:
与现有技术相比,本公开的有益效果是:通过可视化验证该方法的有效性,克服了使用系统识别方法难以验证静态增益估计的问题。
构成本申请的一部分的说明书附图用来提供对本申请的进一步理解,本申请的示意性实施例及其说明用于解释本申请,并不构成对本申请的不当限定。
图1为本发明的基于工业大数中斜坡响应的动态系统静态增益估计流程图;
图2(a)-图2(d)为本发明具体实施示例中的采样数据样本图;
图3(a)-图3(f)为本发明具体实施示例中的计算数据图。
应该指出,以下详细说明都是示例性的,旨在对本申请提供进一步的说明。除非另有指明,本文使用的所有技术和科学术语具有与本申请所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是,这里所使用的术语仅是为了描述具体实施方式,而非意图限制根据本申请的示例性实施方式。如在这里所使用的,除非上下文另外明确指出,否则单数形式也意图包括复数形式,此外,还应当理解的是,当在本说明书中使用术语“包含”和/或“包括”时,其指明存在特征、步骤、操作、器件、组件和/或它们的组合。
专业术语解释:
所述动态系统,是指状态随时间而变化的系统。所述静态增益,是指从一个稳态到新的稳态,系统的单位变化程度。
所述斜坡响应,是指一个输入量的变化斜率从零跃增到某有限值引起的时间响应。
所述输出时间序列,是指输出数值按其发生的时间先后顺序排列而成的数列。
所述输入时间序列,是指输入变量值按其发生的时间先后顺序排列而成的数列。
实施例一
如图1所示,基于历史数据斜坡响应的动态系统静态增益估计方法,包括:
步骤1:将输出y和输入u
i的时间序列分割成短数据段,每个数据段用一条直线表示。我们采用分段线性表示方法描述输出y的分段。
用线性回归模型来描述:
y(n)=a
m+b
mn+e(n),
其中,a
m表示第m段数据段的初始值,b
m表示第m段数据段的斜率,e(n)是干扰;
其中,L(M)为拟合误差损失函数:
步骤2:从数据段中找出输入和输出同时处于直线上且振幅变化较大的斜坡响应。
引入指示序列:
其中,A
y,0为y显著幅度变化的阈值。
步骤3:由输入和输出有显著振幅变化的斜坡响应估计静态增益。
从K的高斯分布,可以估计K的置信区间。
步骤3.4:此时,剩余的具有显著变化幅值的数据集合为:
本实施例中各步骤的结果都可以通过目测相关图形来验证准确性。例如,查找到的斜坡响应,可以通过检查输入和输出数据段是否在直线上来验证查找的准确性。
实施例二
本实施例的目的是提供一种大型300MW燃煤发电机组静态增益估计方法。
为了实现上述目的,本实施例提供了一种大型300MW燃煤发电机组静态增益估计方法,包括以下步骤:
采集机组中产生的有功功率作为输出,采集主蒸汽流量的控制器输出和主蒸汽压力作为输入,得到数据样本;
将输出和输入的时间序列分割成短数据段,每个数据段用一条直线表示;
从数据段中找出输入和输出同时处于直线上且振幅变化较大的斜坡响应;
由输入和输出有显著振幅变化的斜坡响应估计燃煤发电机组的静态增益。
以下是本发明所述方法在具体示例中的应用。
以某大型300MW燃煤发电机组为例,在采样周期h=1s的情况下,采集了机组中的有功功率(y)数据样本作为系统输出、主蒸汽流量的控制器输出(u
1)和主蒸汽压力(u
2)数据样本作为系统输入。估计u
1和y的静态增益G
1及u
2和y的静态增益G
2。
第一步,在2018年5月31日,采用分段线性表示方法对一个小时的数据样本进行分段。图2(a)和图2(b)为
的时序图,计算出
表示将一小时数据样本分为3段。类似地,可以得到了
和
的分段结果,如图2(c)和图2(d)所示。
表1为输出y和输入u
1、u
2的数据段振幅变化量。对于图2(a)中y的三个数据段,其振幅的变化都大于A
y,0;对于图2(c)中的u
1,前三个振幅变化大于
对于图2(d)中的u
2,五个振幅变化中只有两个值大于
根据总指示序列I(n),得到三段输入和输出同时有显著振幅变化的数据段:[1,658],[1136,1677]和[1678,2305]。得到
第三步,从集合
中估计静态增益组,如图3(a)、图3(b)和图3(c)所示,选择A
y,l和
之间最大偏差的可接受水平
表2给出了三组静态增益估计及其置信区间。如图3(d)、图3(e)和图3(f)所示A
y,l和
之间偏差都小于δ
y。说明此方法估计静态增益的有效性。
表1本发明具体实施示例中样本数据信息表
表2本发明具体实施示例中静态增益计算结果和置信区间表
| 组别 | 静态增益 | 估计值 | 置信区间 |
| #1 | K 1 | 1.8781 | [1.8452,1.9110] |
| K 2 | 18.9280 | [18.6216,19.2344] | |
| #2 | K 1 | 2.3132 | [2.2720,2.3544] |
| K 2 | 22.7161 | [22.2615,23.1707] | |
| #3 | K 1 | 2.2287 | [2.1840,2.2735] |
| K 2 | 15.7881 | [15.3865,16.1896] |
实施例三
本实施例的目的是基于实施例二所述静态增益估计方法,提供了一种反馈控制器设计方法。具体包括以下步骤:
采集机组中产生的有功功率作为输出,采集主蒸汽流量的控制器输出和主蒸汽压力作为输入,得到数据样本;
将输出和输入的时间序列分割成短数据段,每个数据段用一条直线表示;
从数据段中找出输入和输出同时处于直线上且振幅变化较大的斜坡响应;
由输入和输出有显著振幅变化的斜坡响应估计燃煤发电机组的静态增益;
将所述静态增益用于所述发电机组的反馈控制器设计。
以设计一种大型300MW燃煤发电机组的某一阶加延迟系统的反馈控制器为例,反馈控制器C(s)限定为PI形式,那么反馈控制器可以表示为:
假设一阶加延迟系统
的时间常数T和延迟τ已知。所述反馈控制器设计方法包括以下步骤:
采集其相关的输出(y)、和输入(u)的数据样本;
将输出y和输入u的时间序列分割成短数据段,每个数据段用一条直线表示;
从数据段中找出输入u和输出y同时处于直线上且振幅变化较大的斜坡响应;
由有显著振幅变化的输入和输出的斜坡响应估计出动态系统的静态增益K;由已知的系统时间常数T和延迟τ,及求得的静态增益K;使用Ziegler-Nichols参数整定法,根据经验整定公式,计算出反馈控制器参数K
p和T
i;
上述实施例二和三中涉及的各步骤均与实施例一相对应,具体实现方法可参照实施例一。
以上所述仅为本申请的优选实施例而已,并不用于限制本申请,对于本领域的技术人员来说,本申请可以有各种更改和变化。凡在本申请的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本申请的保护范围之内。
Claims (7)
- 基于历史数据斜坡响应的动态系统静态增益估计方法,其特征是,包括:采用分段线性表示方法将动态系统的输出时间序列分割成若干个输出时间段;同时,将动态系统的输入时间序列分割成若干个输入数据段;每个数据段均用一条直线表示;找出输入值和输出值同时均处于各自的直线数据段上且输入值和输出值的振幅变化均超过设定阈值的数据段,此数据段即为目标斜坡响应,将找出的所有输入数据段和所有输出数据段组成具有显著振幅变化的数据段集合;通过具有显著振幅变化的数据段集合,估计静态增益。
- 如权利要求1所述的方法,其特征是,找出输入值和输出值同时均处于各自的直线数据段上且输入值和输出值的振幅变化均超过设定阈值的数据段,此数据段即为目标斜坡响应,将找出的所有输入数据段和所有输出数据段组成具有显著振幅变化的数据段集合的具体步骤为:对于输出来说,计算第m段的振幅变化量:引入序列:其中,A y,0为y显著幅度变化的阈值;整体序列:
- 如权利要求1所述的方法,其特征是,通过具有显著振幅变化的数据段集合,估计静态增益的具体步骤为:从K的高斯分布,估计K的置信区间;步骤3.4:剩余的具有显著变化幅值的数据段集合为:
- 一种大型300MW燃煤发电机组静态增益估计方法,其特征在于,采集机组中产生的有功功率作为输出,采集主蒸汽流量的控制器输出和主蒸汽压力作为输入,得到数据样本;将输出和输入的时间序列分割成短数据段,每个数据段用一条直线表示;从数据段中找出输入和输出同时处于直线上且振幅变化较大的斜坡响应;由输入和输出有显著振幅变化的斜坡响应估计燃煤发电机组的静态增益。
- 一种大型300MW燃煤发电机组反馈控制器设计方法,其特征在于,包括以下步骤:采集机组中产生的有功功率作为输出,采集主蒸汽流量的控制器输出和主蒸汽压力作为输入,得到数据样本;将输出和输入的时间序列分割成短数据段,每个数据段用一条直线表示;从数据段中找出输入和输出同时处于直线上且振幅变化较大的斜坡响应;由输入和输出有显著振幅变化的斜坡响应估计燃煤发电机组的静态增益;将所述静态增益用于所述发电机组的反馈控制器设计。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA3109182A CA3109182C (en) | 2018-12-12 | 2019-04-30 | Static gain estimation method for dynamic system based on historical data ramp responses |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811517302.0 | 2018-12-12 | ||
| CN201811517302.0A CN109635431B (zh) | 2018-12-12 | 2018-12-12 | 基于历史数据斜坡响应的动态系统静态增益估计方法 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020119012A1 true WO2020119012A1 (zh) | 2020-06-18 |
Family
ID=66073027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2019/085188 Ceased WO2020119012A1 (zh) | 2018-12-12 | 2019-04-30 | 基于历史数据斜坡响应的动态系统静态增益估计方法 |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN109635431B (zh) |
| CA (1) | CA3109182C (zh) |
| WO (1) | WO2020119012A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112749496A (zh) * | 2020-11-25 | 2021-05-04 | 中国人民解放军国防科技大学 | 基于时序作战环的装备体系作战效能评估方法及系统 |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109635431B (zh) * | 2018-12-12 | 2019-10-29 | 山东科技大学 | 基于历史数据斜坡响应的动态系统静态增益估计方法 |
| CN110350604B (zh) * | 2019-07-18 | 2020-12-01 | 国网山东省电力公司电力科学研究院 | 基于静态特性的火电机组一次调频指标估计方法 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102012017A (zh) * | 2010-11-19 | 2011-04-13 | 华北电力大学(保定) | 一种锅炉汽温自动控制系统中前馈信号控制方法 |
| CN105275508A (zh) * | 2015-11-06 | 2016-01-27 | 国网河南省电力公司电力科学研究院 | 一种基于功率值计算的汽轮机流量曲线辨识及优化方法 |
| US20170058715A1 (en) * | 2015-08-28 | 2017-03-02 | General Electric Company | Control system for managing steam turbine rotor stress and method of use |
| CN108549346A (zh) * | 2018-05-14 | 2018-09-18 | 山东科技大学 | 一种适于系统辨识的历史数据段自动查找方法 |
| CN109635431A (zh) * | 2018-12-12 | 2019-04-16 | 山东科技大学 | 基于历史数据斜坡响应的动态系统静态增益估计方法 |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6684115B1 (en) * | 2000-04-11 | 2004-01-27 | George Shu-Xing Cheng | Model-free adaptive control of quality variables |
| JP4422179B2 (ja) * | 2007-10-19 | 2010-02-24 | 株式会社半導体理工学研究センター | 半導体集積回路のタイミング解析装置及び方法 |
| GB0721909D0 (en) * | 2007-11-08 | 2007-12-19 | Cnh Belgium Nv | Apparatus and method for controlling the speed of a combine harvester |
| CN101877025A (zh) * | 2010-07-02 | 2010-11-03 | 天津大学 | 复杂非线性静态特性描述的分布式电源模型简化方法 |
| CN102778844B (zh) * | 2012-07-30 | 2014-08-13 | 杭州电子科技大学 | 基于有限元模型和系统辨识的感应加热闭环仿真方法 |
| CN102779216B (zh) * | 2012-07-30 | 2014-09-17 | 杭州电子科技大学 | 基于有限元模型的电磁感应加热过程系统辨识方法 |
| US9249751B2 (en) * | 2013-05-23 | 2016-02-02 | Ford Global Technologies, Llc | Exhaust gas sensor controls adaptation for asymmetric degradation responses |
| CN103744286A (zh) * | 2013-12-31 | 2014-04-23 | 广东电网公司电力科学研究院 | 一种火力发电系统的控制器的设计方法和装置 |
| CN108205311B (zh) * | 2018-01-14 | 2020-12-18 | 山东科技大学 | 一类事件触发传输时变系统基于未知输入观测器技术的故障估计方法 |
-
2018
- 2018-12-12 CN CN201811517302.0A patent/CN109635431B/zh active Active
-
2019
- 2019-04-30 CA CA3109182A patent/CA3109182C/en active Active
- 2019-04-30 WO PCT/CN2019/085188 patent/WO2020119012A1/zh not_active Ceased
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102012017A (zh) * | 2010-11-19 | 2011-04-13 | 华北电力大学(保定) | 一种锅炉汽温自动控制系统中前馈信号控制方法 |
| US20170058715A1 (en) * | 2015-08-28 | 2017-03-02 | General Electric Company | Control system for managing steam turbine rotor stress and method of use |
| CN105275508A (zh) * | 2015-11-06 | 2016-01-27 | 国网河南省电力公司电力科学研究院 | 一种基于功率值计算的汽轮机流量曲线辨识及优化方法 |
| CN108549346A (zh) * | 2018-05-14 | 2018-09-18 | 山东科技大学 | 一种适于系统辨识的历史数据段自动查找方法 |
| CN109635431A (zh) * | 2018-12-12 | 2019-04-16 | 山东科技大学 | 基于历史数据斜坡响应的动态系统静态增益估计方法 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112749496A (zh) * | 2020-11-25 | 2021-05-04 | 中国人民解放军国防科技大学 | 基于时序作战环的装备体系作战效能评估方法及系统 |
| CN112749496B (zh) * | 2020-11-25 | 2022-09-27 | 中国人民解放军国防科技大学 | 基于时序作战环的装备体系作战效能评估方法及系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| CA3109182A1 (en) | 2020-06-18 |
| CN109635431A (zh) | 2019-04-16 |
| CN109635431B (zh) | 2019-10-29 |
| CA3109182C (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106933992B (zh) | 基于数据分析的分布式数据清洗系统及方法 | |
| CN109753634B (zh) | 基于历史数据稳态值的动态系统增益估计方法 | |
| WO2020119012A1 (zh) | 基于历史数据斜坡响应的动态系统静态增益估计方法 | |
| Chan et al. | TOWARD A UNIFIED INTERVAL ESTIMATION OFAUTOREGRESSIONS | |
| JP2015011027A (ja) | 時系列データにおける異常を検出する方法 | |
| CN117057513A (zh) | 一种基于互联网的智慧园区用监控管理系统 | |
| CN109558968B (zh) | 风电场出力相关性分析方法及装置 | |
| CN108415884B (zh) | 一种结构模态参数实时追踪方法 | |
| CN106202977B (zh) | 一种基于盲源分离算法的低频振荡模式分析方法 | |
| CN108205713B (zh) | 一种区域风电功率预测误差分布确定方法和装置 | |
| CN110442911B (zh) | 一种基于统计机器学习的高维复杂系统不确定性分析方法 | |
| Zhang et al. | A least squares estimator for discretely observed Ornstein–Uhlenbeck processes driven by symmetric α-stable motions | |
| CN105912822B (zh) | 海水脱硫装置脱硫效率的神经网络模拟方法 | |
| CN105931130A (zh) | 一种考虑量测信号丢失的改进集合卡尔曼滤波估计方法 | |
| CN112766535A (zh) | 一种考虑负荷曲线特征的建筑负荷预测方法及系统 | |
| CN118708995B (zh) | 基于人工智能的核电工程调试废水处理监管方法及系统 | |
| CN110705099A (zh) | 一种检定风电场出力相关性的方法 | |
| Bacci di Capaci et al. | Stiction quantification: A robust methodology for valve monitoring and maintenance scheduling | |
| CN109818377B (zh) | 一种基于振幅变化的自动发电控制性能评估方法及系统 | |
| Zajic et al. | Modelling of an air handling unit: a Hammerstein-bilinear model identification approach | |
| CN111127242A (zh) | 一种基于小样本数据的电力系统可靠性动态实时评估方法 | |
| CN111709350A (zh) | 基于fcm聚类的低频振荡模态参数识别方法及系统 | |
| US10805133B2 (en) | Method and apparatus for determining peak power, peak-to-average power ratio | |
| CN108363679B (zh) | 一种自动追踪结构模态参数的方法 | |
| CN105656453B (zh) | 一种基于时间序列的光纤电流互感器随机噪声实时滤波方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| ENP | Entry into the national phase |
Ref document number: 3109182 Country of ref document: CA |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 19896372 Country of ref document: EP Kind code of ref document: A1 |






