Features Research & Innovation
Proven Tools To Measure And Control Energy Use

An analysis of the plant’s energy use was the jumping off point for most of the real-life energy conservation examples presented at the Quebec Forest Industry Council’s (CIFQ) energy conference held last spring in Bcancour, Que.

Cascades Groupe’s two-stage approach to cutting down on energy consumption for plant heating was described by Andr-Anne De Gagn, an engineer with Cascades nergie. The first stage involved determining the real, or baseline, cost of heating in their pulp and paper mills. It was a two-step process that examined the method of calculating energy used for heating and the problems of distinguishing the energy used for heating from overall mill energy consumption.

The second stage then implemented measures to reduce heating costs after an examination of the available technologies for this purpose.

The first problem to resolve in establishing overall mill energy consumption was to determine a method or model to isolate the impact on energy consumption caused by external ambient air temperature fluctuations. The model was based on the total of three variables: production demand plus degree-days of heating plus the base energy load. Calculations demonstrated that for a paper mill, heating represented a cost of $4 to $13 per tonne of product and from 2% to 10% of the overall energy costs for the mill. In a conversion plant, heating costs represented 10% to 15% of total energy costs.

In one example, De Gagn explained that the automation of the Cascades Jonquire mill economizers at a cost of $65,000 yielded savings of $35,000 for ROI of 1.9 years. It also avoided costs of $170,000 per year.

De Gagn also discussed heating by glycol at Cascade Papers Kingsey Falls. An expenditure of $215,000 led to savings of $85,000, for an ROI of 2.5 years. Another example was the installation of natural-gas powered, infrared radiant heaters in another mill to replace electric, steam, and standard natural gas heaters and fans.

Energy management minimizes electricity purchases

P.A. Bessette, an advanced control specialist with SFK Pulp of St. Flicien, Que., described the advanced controls used at the plant in kraft pulp production, with a unique twist — electricity demand management.

SFK uses advanced controls in six applications. From an energy management point of view, the turbo-generator advanced control setup installed in 2007 is probably the most intriguing. The principal goal was to maximize utilization of the turbo-generators, minimize the use of make-up bunker oil, and maximize the profitability of mill electricity production. The mill’s strategy was to limit total outside purchases of electricity as a function of its capacity to produce steam.

SFK Pulp had a cogeneration contract with Hydro-Qubec, whereby the mill sold 15% of electricity from Turbo #1 and 100% from Turbo #2. It purchased electricity from Hydro-Qubec under a “sale plus total consumption less internal production” arrangement subject to three tariff levels: Tariff L for 5 MW, Tariff H for off-peak demand (nights and weekends), and Tariff H for weekday peak demand — quite a headache for the operating team to optimize! The trick was to balance the supply of electricity from both turbo-generators to satisfy mill demand, while optimizing electricity purchases from Hydro-Qubec under the complex tariff terms of the contract.

The mill was able to achieve this and provide the operator with an interface for the steam network in the control room that confirmed the actual cost of power on a continuous real time basis.

Data Mining: Seeking The Big Picture

A further approach to mill energy management using data mining has been adopted at AbitibiBowater. As explained by Sebastien Lafourcade of Pepite Technologies and Martin Fairbanks of AbitibiBowater, data mining is the practice of analyzing process information to reveal process performance information and to generate understanding. In this case, the companies were able to extract added value from historical process data already available to AbitibiBowater, to provide different levels in the mills with better decision making tools for energy management, with little or no financial investment.

The production process can involve more than 3,000 variables. The challenge is to reduce this to fewer than 10 key process variables through preliminary analysis, data exploration, invariate and multivariate analysis, modelization, and advanced multivariate analysis.

At AbitibiBowater, the benefits were substantial: in the first mill they were able to optimize the quantity of dirty TMP steam sent to the regenerator with potential savings of $1 to $4 per tonne. Using an intelligent alarm to track performance of the TMP regenerator, they determined that the uncorrected anomalies they identified were costing $1,000 a day. In a second mill, they observed wide variations in specific steam consumption for a paper machine with three specific consumption peaks, revealing potential savings of $3 to $6 per tonne.

These examples demonstrate that there are efficiencies to be found in mills, and that, once found, these problem areas can be transformed to real savings that flow directly to the bottom line.