【正文】 ative large), THEN the output is given by the local CARIMA model i: .. .)()(?.. .)1(?)(? 01 ????????? dtubntyatyaty iiaiiniiia )(.. .)()( ciinibiin ntectendtubcb ???????? (4) or )()()()()(?)( 111 tezCtuzBztyzA iiiidii ???? ??? (5) Where )()(),( 111 ??? zandCzBzA iii are polynomials in the backward shift operator 1?z , and d is the dead time of the plant, )(tui is the control, and )(tei is a zero mean independent random variable with a variance of 2? . The multivariate basis function )(ki xa is obtained by the tensor products of the univariate basis functions, 17 pixAa nk kiki ,...,2,1,)(1 ???? ? (6) where n is the dimension of the input vector x, and p, the total number of weights in the NFN, is given by, ?? ??nk ii kRp 1 )( (7) Where ik and iR are the order of the basis function and the number of inner knots respectively. The properties of the univariate Bspline basis functions described previously also apply to the multivariate basis function, which is defined on the hyperrectangles. The output of the NFN is, ?????? ??piiipiipiiiayaayy111 ??? (8) 3. Neurofuzzy modelling and predictive control of superheated steam temperature Let? be the superheated steam temperature, and ?? , the flow of spray water to the high temperature superheater. The response of? can be approximated by a second order model: sp esTsT KssG ??? ? ????? )1)(1()()( 21 (9) The linear models, however, only a local model for the selected operating point. Since load is the unique antecedent variable, it is used to select the division between the local regions in the NFN. Based on this approach, the load is divided into five regions as shown in ,using also the experience of the operators, who regard a load of 200MW as high,180MW as medium high,160MW as medium,140MW as medium low and 120MW as low. For a sampling interval of 30s, the estimated linear local models )( 1?zA used in the NFN are shown in Table 1. 18 Fig. 3 Membership function for local models Table 1 Local CARIMA models in neurofuzzy model Cascade control scheme is widely used to control the superheated steam temperature. Feed forward control, with the steam flow and the gas temperature as inputs, can be applied to provide a faster response to large variations in these two variables. In practice, the feed forward paths are activated only when there are significant changes in these variables. The control scheme also prevents the faster dynamics of the plant, ., the spray water valve and the water/steam mixing, from affecting the slower dynamics of the plant, ., the high temperature superheater. With the global nonlinear NFN model in Table 1, the proposed NFGPC scheme is shown in . 19 Fig. 4 NFGPC control of superheated steam temperature with feedforward control. As a further illustration, the power plant is simulated using the NFN model given in Table 1,and is controlled respectively by the NFGPC, the conventional linear GPC controller, and the cascaded PI controller while the load changes from 160MW to conventional linear GPC controller is the local controller designed for the“ medium” operating region. The results are shown in ,showing that, as expected, the best performance is obtained from the NFGPC as it is designed based on a more accurate process model. This is followed by the conventional linear GPC controller. The performance of the conventional cascade PI controller is the worst, indicating that it is unable to control satisfactory the superheated steam temperature under large load changes. This may be the reason for controlling the power plant manually when there are large load changes. parison of the NFGPC, conventional linear GPC, and cascade PI controller. 4. Conclusions The modeling and control of a 200 MW power plant using the neurofuzzy approach is presented in this paper. The NFN consists of five local CARIMA models. 20 The output of the work is the interpolation of the local models using memberships given by the Bspline basis functions. The proposed NFGPC is similarly constructed, which is designed from the CARIMA models in the NFN. The NFGPC is most suitable for processes with smooth nonlinearity, such that its full operating range can be partitioned into several local linear operating regions. The proposed NFGPC therefore provides a useful alternative for controlling this class of nonlinear power plants, which are formerly difficult to be controlled using traditional methods. 21 Part 4 為 Part3 譯文： 鍋爐蒸汽溫度模糊神經(jīng)網(wǎng)絡(luò)的廣義預(yù)測控制 摘要 ： 發(fā)電廠是非線性和不確定 性 的復(fù)雜 系統(tǒng) ?，F(xiàn)代電廠的運行中，為確保電廠的高效率和高負荷的能力，準確的 控制過熱蒸汽溫度是必要的。本文提出 了一類 在 非線性廣義預(yù)測控制器的基礎(chǔ)上 的 模糊神經(jīng)網(wǎng)絡(luò) 。所提出的 非線性控制器適用于控制一臺 200 MW 電廠 的 過熱蒸汽溫度。 從實驗方案的仿真結(jié)果中可以看出，此方案的控制品質(zhì)優(yōu)于 傳統(tǒng)的控制 方案。 關(guān)鍵詞 ：模糊神 經(jīng)網(wǎng)絡(luò) ； 廣義預(yù)測控制 ； 過熱蒸汽溫度 1. 引言 電廠過熱汽溫控制 系統(tǒng)的特點是非線性 、 不確定性和負載擾動。蒸汽發(fā)電的過程中鍋爐 汽輪機 溫度 過熱是一個重要的 問題 ，蒸汽 加熱后 ，進入渦輪驅(qū)動發(fā)電機 ， 控制過熱蒸汽溫度不僅是 在技術(shù)上具有挑戰(zhàn)性，在經(jīng)濟上 的意義 也是 十分重要的。 圖 1 鍋爐過熱器和蒸汽生成過程 從圖 1可以看出， 產(chǎn)生的蒸汽從鍋爐汽包通過低溫過熱器 后 進入輻射型屏。水 變成 噴涂的蒸汽，以控制過熱蒸汽 的 溫度。適當?shù)目刂齐姀S過熱蒸汽溫度是極其重要的， 可以 確保整體效率和安全性。蒸汽溫度太高 對系統(tǒng)是非常不利的 ，因為過熱蒸汽 可以 損害高壓力汽輪機，太低 也不行 ，因為它會降低電廠 熱 效率。減少溫度波動也是 非常 重要的，因為它有助于減少在單位內(nèi)機械應(yīng)力造成的微裂紋，延長單位秩序壽命，并減少維修成本。作為 GPC的推導(dǎo) 應(yīng)該 盡量減少這些波動，它是 眾多的控制器 中 最適合實現(xiàn)這一目標的。 22 多變量多步自適應(yīng)調(diào)節(jié)已適用于控制過熱蒸汽溫度在 150 ht/ 的 鍋爐 ，提出了 廣義預(yù)測控制以控制蒸汽溫度 ， 基于神經(jīng)網(wǎng)絡(luò)發(fā)展 的 非線性預(yù)測控制器是以控制主蒸汽溫度和壓力 。 控制主蒸汽 溫度和 壓力的基礎(chǔ)上， 得到 非線性模型的構(gòu)成是 由 非線性靜力常數(shù)和非線性動力學(xué) 組成 。 模糊邏輯是把人類的經(jīng)驗透過模糊規(guī)則表現(xiàn)出來。然而，設(shè)計模糊邏輯控制器是 非常消費 時間 的 ，由于模糊規(guī)則 的不確定， 往往得到的試驗是錯誤的。在此相反，神經(jīng)網(wǎng)絡(luò)不僅有近似的非線性職能與任意精度，他們也可以有 經(jīng)過試驗 的實驗數(shù)據(jù)。該模糊神經(jīng)網(wǎng)絡(luò) 的 開發(fā)優(yōu)勢 是